This page intentionally left blank Lichen Biology
Lichens are symbiotic organisms in which fungi and algae and/or cyanobacteria form an intimate biological union. This diverse group is found in almost all terrestrial habitats from the tropics to polar regions. In this second edition, four completely new chapters cover recent developments in the study of these fascinating organisms, including lichen genetics and sexual reproduction, stress physiology and symbiosis, and the carbon economy and environmental role of lichens. The whole text has been fully updated, with chapters covering anato- mical, morphological and developmental aspects; the chemistry of the unique secondary metabolites produced by lichens and the contribution of these sub- stances to medicine and the pharmaceutical industry; patterns of lichen photosynthesis and respiration in relation to different environmental condi- tions; the role of lichens in nitrogen fixation and mineral cycling; geographical patterns exhibited by these widespread symbionts; and the use of lichens as indicators of air pollution. This is a valuable reference for both students and researchers interested in lichenology.
T H O M A S H . N A S H I I I is Professor of Plant Biology in the School of Life Sciences at Arizona State University. He has over 35 years teaching experience in Ecology, Lichenology and Statistics, and has taught in Austria (Fulbright Fellowship) and conducted research in Australia, Germany (junior and senior von Humbolt Foundation fellowships), Mexico and South America, and the USA. He has coauthored 5 previous books and over 200 scientific articles.
Lichen Biology
Second Edition
Edited by THOMAS H. NASH III Arizona State University, USA CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521871624
© Cambridge University Press 1996, 2008
This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2008
ISBN-13 978-0-511-41407-7 eBook (EBL)
ISBN-13 978-0-521-87162-4 hardback
ISBN-13 978-0-521-69216-8 paperback
Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Contents
List of contri butorsvii Preface ix
1 Introduction 1 T. H. NASH III
2 Photobionts 9 T. FRIEDL AND B. BU¨ DEL
3 Mycobionts 27 R. HONEGGER
4 Thallus morphology and anatomy 40 B. BU¨ DEL AND C. SCHEIDEGGER
5 Morphogenesis 69 R. HONEGGER
6 Sexual reproduction in lichen-forming ascomycetes 94 R. HONEGGER AND S. SCHERRER
7 Biochemistry and secondary metabolites 104 J. A. ELIX AND E. STOCKER-WO¨ RGO¨ TTER
8 Stress physiology and the symbiosis 134 R. P. BECKETT,I.KRANNER, AND F. V. MINIBAYEVA
9 Physiological ecology of carbon dioxide exchange 152 T. G. A. GREEN,T.H.NASH III, AND O. L. LANGE
v vi Contents
10 The carbon economy of lichens 182 K. PALMQVIST,L.DAHLMAN,A.JONSSON, AND T. H. NASH III
11 Nitrogen, its metabolism and potential contribution to ecosystems 216 T. H. NASH III
12 Nutrients, elemental accumulation, and mineral cycling 234 T. H. NASH III
13 Individuals and populations of lichens 252 D. FAHSELT
14 Environmental role of lichens 274 M. R. D. SEAWARD
15 Lichen sensitivity to air pollution 299 T. H. NASH III
16 Lichen biogeography 315 D. J. GALLOWAY
17 Systematics of lichenized fungi 336 A. TEHLER AND M. WEDIN
Appendix: Culture methods for lichens and lichen symbionts 353 E. STOCKER-WO¨ RGO¨ TTER AND A. HAGER
References 364 Taxon index 462 Subject index 477 Contributors
Dr. Richard Beckett School of Biological and Conservation Sciences, University of Kwazulu- Natal, Private Bag X01, Pietermaritzburg, South Africa Professor Dr. Burkhard Bu¨ del Department of General Botany, Department of Biology, Erwin-Schro¨ dinger- Str. 13 University of Kaiserslautern, D-67663 Kaiserslautern, Germany Dr. Lena Dahlman Department of Ecology and Environmental Science, Umea˚University, SE- 90187 Umea˚, Sweden Emeritus Professor J. A. Elix Department of Chemistry, Faculty of Science, Australian National University, Canberra, ACT 0200, Australia Dr. Dianne Fahselt Department of Plant Sciences, University of Western Ontario, London, Ontario N6A 5B7, Canada Professor Dr. Thomas Friedl Department of Experimental Phycology and Sammlung von Algenkulturen, Albrecht-von-Haller-Institute for Plant Sciences, University of Go¨ ttingen, Untere Karspuele 2, 37073 Go¨ ttingen, Germany Dr. David Galloway 16 Farquharson St., Opoho, Dunedin, New Zealand Dr. T. G. A. Green Department of Biological Sciences, The University of Waikato, Private Bag 3105, Hamilton 3240, New Zealand Professor Dr. Rosmarie Honegger Institute of Plant Biology, University of Zu¨ rich, Zollikerstrasse 107, CH-8008 Zu¨ rich, Switzerland
vii viii List of contributors
Dr. Anna Jonsson Department of Ecology and Environmental Science, Umea˚University, SE-90187 Umea˚, Sweden Dr. Ilse Kranner Seed Conservation Department, Royal Botanic Gardens, Kew, Wakehurst Place, West Sussex, RH17 6TN, UK Professor Dr. Otto L. Lange Julius-von-Sachs-Institute of Biosciences, University of Wuerzburg, Lehrstuhl fuer Botanik II, Julius-von-Sachs-Platz 3, D-97082 Wuerzburg, Germany Dr. Farida Minibayeva Institute of Biochemistry and Biophysics, P.O.Box 30, Kazan, 420111, Russia Dr. Thomas H. Nash III School of Life Sciences, Arizona State University, Box 874501, Tempe, AZ 85287-4501, USA Professor Kristin Palmqvist Department of Ecology and Environmental Science, Umea˚University, SE-90187 Umea˚, Sweden Dr. Christoph Scheidegger Swiss Federal Institute for Forest, Snow and Landscape Research, WSL, Zu¨ rcherstr. 111, CH-8903 Birmensdorf, Switzerland Dr. S. Scherrer Institute of Plant Biology, University of Zu¨ rich, Zollikerstrasse 107, CH-8008 Zu¨ rich, Switzerland Professor Mark R. D. Seaward Department of Geography and Environmental Science, University of Bradford, Bradford, BD7 1DP, UK Professor Dr. Elfie Stocker-Wo¨ rgo¨ tter Department of Organismic Biology (Plant Physiology), University of Salzburg, Hellbrunner Str. 34, A-5020 Salzburg, Austria Dr. Anders Tehler Swedish Museum of Natural History, Box 50007, SE 104 05 Stockholm, Sweden Dr. Mats Wedin Swedish Museum of Natural History, Box 50007, SE 104 05 Stockholm, Sweden Preface to the second edition
Twelve years ago the first edition of Lichen Biology was published, and brought a new synthesis to the field of lichenology. In the meantime, rapid advances in many areas, particularly in molecular biology, have expanded our horizons and added depth to our knowledge of areas already under investigation. Consequently, it is appropriate that a second edition has now been consummated. The original edition had 13 chapters, but this edition has 17 chapters and has added an appendix on lichen culturing, which is becoming prominent in the expanding biotechnology area. New chapters include one on sexual reproduc- tion (Chapter 6), summarizing knowledge not available in 1996. As prominent examples of stress-tolerant organisms, lichens have developed a variety of strategies that allow them to occupy both extremely cold and hot environments; consequently, these investigations were meritorious of a chapter of their own (Chapter 8). In addition, a chapter on growth (Chapter 10), a topic briefly covered in the original photosynthesis chapter, is now expanded to cover much new information and the major advances over the past decade. Although many aspects of the ecology of lichens were covered in the first edition, a number of important areas were omitted. This has been rectified in Chapter 14. Of the remaining chapters, the chapter titles remain the same from the first edition, but all chapters have been revised to a greater or lesser degree. For example, the chapter on the individual (Chapter 13) and air pollution (Chapter 15) bear little resemblance to their original counterparts. Altogether 13 additional people have contributed substantially to this edition. As with the first edition, this book should be of interest to the specialist, whether amateur or professional lichenologist. Furthermore, the book will pro- vide an essential reference for many other people, such as anyone interested in the phenomenon of symbiosis, ecologists interested in the role of lichens in ecosystems, or a land manager charged with assessing the effects of air pollution on natural systems. We also hope it will stimulate the next generation of students and young scientists to advance our knowledge of these wonderful organisms.
ix
1 Introduction
t. h. nash iii
1.1 The symbiosis
Lichens are by definition symbiotic organisms, usually composed of a fungal partner, the mycobiont (Chapter 3), and one or more photosynthetic partners, the photobiont (Chapter 2), which is most often either a green alga or cyanobacterium. Although the dual nature of most lichens is now widely recognized, it is less commonly known that some lichens are symbioses invol- ving three (tripartite lichens) or more partners. The potential relationships of mycobionts and photobionts may in fact be quite complex (Chapter 4), and a rigorous classification of many types of relationships was developed by Rambold and Triebel (1992). In general, lichens exist as discrete thalli and are implicitly treated as individuals in many studies (but see Chapter 13), even though they may be a symbiotic entity involving three kingdoms! From a genetic and evolutionary perspective, lichens can certainly not be regarded as individuals and this fact has major implications for many areas of investigation, such as developmental and reproductive studies (Chapter 5). The nature of the lichen symbiosis is widely debated and deserves further investigation. Most general textbooks and many researchers refer to lichens as a classical case of mutualism, where all the partners gain benefits from the association. Alternatively, lichens are regarded as an example of controlled parasitism, because the fungus seems to obtain most of the benefits and the photobiont may grow more slowly in the lichenized state than when free-living (Ahmadjian 1993). In fact, the relationships may be much more complex, espe- cially when additional lichenicolous fungi (Lawrey and Diederich 2003) occur on/in lichens. These are different fungi from the dominant mycobiont, and they may have a parasitic, commensalistic, mutualistic or saprophytic/saprobic
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. 2 T. H. Nash III
relationship to the lichen (Rambold and Triebel 1992). Parasitic symbiotic fungi may cause extensive damage, resulting in localized necrotic patches or in complete death of the thallus. On the other hand, commensalistic symbiotic fungi apparently share the photosynthetically derived products from the photo- biont with the mycobiont of the existing symbiosis. Such secondary fungi are assumed not to benefit their hosts, and they do not appear to damage them, although they may lead to the formation of gall-like growths, the morphology and physiology of which are, as yet, little understood. A few cases are being discovered where the secondary parasitizing fungus progressively eliminates the primary mycobiont, taking over the photobiont to produce a new thallus of its own; the stability of such a union can be fragile, since in some cases it has been discovered that the original photobiont can be exchanged for another preferred photobiont after the takeover. Within the realm of what we call lichens, the degree of lichenization varies tremendously from a few photobiont cells that seem to be almost haphazardly associated with a fungus (e.g. some Caliciales) to the more typical well-integrated thallus, in which a distinct photobiont layer is found beneath cortical fungal tissue (Chapters 4 and 5). In most of the latter cases, the lichen bears little morphological similarity to the bionts that form it. Because of the differences in degree of lichenization, no single definition may adequately cover the full range of relationships found within lichens. The morphology of the lichenized thallus is strongly influenced by the photobiont and its direct contact with the mycobiont (Chapters 4 and 5). In nature there are at least a few cases where the same mycobiont, as ascertained with molecular techniques, is able to form two very different, interconnected thalli with respectively a cyanobacterium and a green alga (Armaleo and Clerc 1990). These different morphotypes are called photosymbiodemes, and their occurrence implies ontogenetic control by the photobiont. In culture, unliche- nized mycobionts remain relatively amorphous, but initiate thallus develop- ment when they first come in contact with their photobiont (Ahmadjian 1993; Chapter 5). Subsequently the mycobiont may completely envelop the photo- bionts and, particularly in the case of green algae, penetrate the surface of the photobiont with structures called haustoria. Because haustoria are sometimes associated with dead photobiont cells and because parasitic fungi frequently form haustoria, Ahmadjian (1993) interprets lichenization as an example of controlled parasitism. Although there is limited experimental evidence, these haustoria are assumed to facilitate carbohydrate transfer from the photo- biont to the mycobiont. In the future it would be interesting to determine whether haustoria can also facilitate nutrient delivery from the mycobiont to the photobiont. Introduction 3
Certainly there is variation in the degree to which the symbiosis is an obli- gate one for the partners involved. The green algal Trebouxia, which occurs in approximately 20% of all lichens, has rarely been found free-living (Chapter 2). In contrast, other photobiont genera, such as Gleocapsa, Nostoc, Scytonema, and Trentepohlia, occur commonly both in lichenized and free-living states. In at least some cases, both free-living and lichenized populations occur in the same habitat, such as free-living Nostoc and Scytonema in desert soils and their lichenized counterparts respectively in the terricolous lichens Collema and Peltula. The degree to which the same photobiont species occurs in both free- living and lichenized states (Beck 2002) is not well established, because rela- tively few lichen algae have been definitively identified to species, and more generally, the systematics at the species level of many cyanobacteria and uni- cellular green algae are not well resolved (Chapter 2). Nevertheless, it appears that most lichens are highly specific in their choice of photobiont (Beck et al. 1998; Rambold et al. 1998). In contrast, the systematics of the mycobiont is well known. Because isolated mycobionts grow so slowly, they are unlikely to survive well in the free-living state due to competition with other fungi or consumption by other organisms. Thus, most mycobionts are assumed to have an obligate relationship to lichenization, although the specificity of the mycobiont for a particular photobiont may not be as great as one might assume. In addition to the photosymbiodeme example cited above, more than one species of Trebouxia have been isolated from the same thallus (Friedl 1989b; Ihda et al. 1993). Overall, the lichen symbiosis is a very successful one as lichens are found in almost all terrestrial habitats from the tropics to polar regions (Chapter 14). Certainly as a result of the symbiosis, both photobiont and mycobiont have expanded into many habitats, where separately they would be rare or non existent. For example, most free-living algae and cyanobacteria occur in aqua- tic or at least very moist terrestrial habitats, but as part of lichens they occur abundantly in habitats that are frequently dry as well. Not only may the fungus enhance water uptake due to its low water potential (see below), but also it substantially reduces the light intensity to which the photobiont is exposed (Ertl 1951). High light intensity adversely affects the photobiont (Demmig- Adams et al. 1990), and hence lichenization is one mechanism by which pho- tobionts may expand into high light environments. Thus, there may well be benefits to lichenization from the perspective of the photobiont. Overall, it may be less important to evaluate lichenization from a strict cost/benefit perspective than to recognize it as a prominent example of a successful sym- biosis. Additional studies will doubtlessly help to elucidate further our under- standing of the symbiosis. 4 T. H. Nash III
1.2 Systematics
Lichens are classified as fungi (Chapter 17), and estimates of the number of species vary from 13 500 (Hawksworth and Hill 1984) to approximately 17 000 (Hale 1974). Because many regions of the world have been poorly collected, the higher number may well be more reasonable. By far the largest number of lichens are Ascomycetes and in fact almost half of the described Ascomycetes are liche- nized (Chapter 17). In addition, there are a few lichenized Basidiomycetes and Deuteromycetes (¼ Fungi Imperfecti). The latter group is an artificial class, in which sterile species are placed. If fruiting structures are eventually found, then these lichens may in due course be classified as either Ascomycetes or Basidio- mycetes. In addition, in the Actinomycetes, Mastigomycetes and Myxomycetes, there are a few symbiotic associations with some properties similar to lichens, but in general these are excluded from lichen classifications. Although one might hypothesize that cyanobacteria, green algae, and fungi evolved from lichens, it is generally assumed that lichenization occurred sub- sequent to development of these organisms. In the fossil record there is limited evidence for the occurrence of lichens, but this may be more due to lack of preservation than their absence from earlier eons. In fact, several quite old fossils have recently been interpreted as being lichens (Chapters 5 and 16). The diversity of lichenized fungi and the fact that some groups contain both lichenized and free-living fungi has led to the inference that lichenization and delichenization have occurred more than once and in fact may have occurred several times (Gargas et al. 1995; Lutzoni et al. 2001). The initial inference is supported by the occurrence of lichens in different classes of fungi, and, within the Ascomycetes, by the fact that lichenization occurs exclusively in only five of the 16 orders, in which lichenization has thus far been found (Hawksworth 1988a). If lichenization has occurred multiple times, then in an evolutionary sense lichens cannot be regarded as one group or, as a phylogeneticist would say, lichens are polyphyletic (Chapter 17).
1.3 Diversity and ecological domain of lichens
Among the terrestrial autotrophs of the world, lichens exhibit intri- guing morphological variation in miniature (Chapter 4). In color they exhibit a fantastic array of orange, yellow, red, green, gray, brown, and black (Wirth 1995; Brodo et al. 2001). Lichens vary in size from less than a mm2 to long, pendulous forms that hang over 2 m from tree branches (Chapter 4). Almost all lichens are perennials, although a few ephemerals (e.g. Vezdaea) are known. At the other extreme some lichens are estimated to survive well over 1000 years and may be Introduction 5 useful in dating rock surfaces (Beschel 1961; Section 10.7). Linear growth varies from imperceptible to many millimeters in a year. Lichens occur commonly as epiphytes on trees and other plants, and in some ecosystems epiphytic lichen biomass may exceed several hundred kg ha 1 (Coxson 1995). In addition, they frequently colonize bare soil, where they are an important component of cryptogamic soil crusts in arid and semi-arid land- scapes (Evans and Johansen 1999; Belnap and Lange 2003). Furthermore, lichens occur almost ubiquitously on rocks with the most obvious ones occurring as epiliths, either growing over the surface or embedded within the upper few millimeters. A few lichens even occur endolithically within the upper few millimeters of the rock, such as occurs in Antarctica (Friedmann 1982). In the tropics and subtropics, some rapidly growing lichens even colonize the surface of leaves as epiphylls (Lu¨ cking and Bernecker-Lu¨ cking 2002). Although most lichens are terrestrial, a few occur in freshwater streams (e.g. Peltigera hydro- thyria) and others occur in the marine intertidal zone (e.g. Lichina spp. and the Verrucaria maura group). Lichens occur in most terrestrial ecosystems of the world, but their bio- mass contribution varies from insignificant to being a major component of the whole ecosystem (Kershaw 1985; Chapter 14). In many polar and subpolar ecosystems, lichens are the dominant autotrophs (Longton 1988). In addition, lichens are conspicuous components of many alpine, coastal and forest ecosys- tems, such as the temperate rain forests of the southern hemisphere (Galloway 2007) and taiga of the northern hemisphere (Kershaw 1985). Because most lichens grow relatively slowly, their primary productivity contribution is fairly small in most ecosystems (Chapter 10). On the other hand, the more rapidly growing species may increase their biomass by 20–40% in a year and these species may play an important role in the mineral cycling patterns of their ecosystems (Section 12.10), particularly if cyanolichens are the dominant com- ponent (Chapter 11).
1.4 Lichens as poikilohydric organisms
Most flowering plants and conifers have developed the capacity to maintain the water status of their leaves or needles at fairly constant levels and hence are referred to as homiohydric organisms. In contrast, lichens are prominent members of poikilohydric organisms, whose water status varies passively with surrounding environmental conditions (Chapter 9). Other poiki- lohydric organisms include the bryophytes, some ferns and other primitive vascular plants. All of these organisms become desiccated relatively rapidly and, as a consequence, water availability is of prime importance for their 6 T. H. Nash III
150 interior
) Coastal –2 31 km inland 100
50 y = 0.135x – 15.019 Lichen biomass (g m
0 50 100 150 200 250 300 Precipitation (mm)
Fig. 1.1 Relationship of biomass of lichen communities within the Sonoran Desert region to mean annual precipitation (Nash and Moser 1982).
survival and in explaining their patterns of occurrence (Chapter 9). One might assume that poikilohydric organisms are highly dependent on precipitation, primarily in the form of rain. Certainly this is true for many lichens, as can be seen for the lichen biomass relationship among interior desert sites (Fig. 1.1, the straight line). On the other hand, lichen biomass near the Pacific Ocean in the western part of the Sonoran Desert vastly exceeds values that would be pre- dicted based on precipitation alone (Fig. 1.1, crosses). This illustrates the ability of lichens to utilize other water sources, such as fog and dew. In addition, lichens have the remarkable ability to extract some moisture from non satu- rated air under conditions of low temperatures and high humidities. This is essentially the reverse of transpirational water flow occurring through vascular plants and is due to the low osmotic values of lichen thalli. However, under intermediate to high temperatures and intermediate to low humidities, the water potential gradient from the lichen to the atmosphere is reversed and evaporation occurs.
1.5 Practical applications
Many of the secondary products formed by lichens are unpalatable and may serve as defensive compounds against herbivores as well as decomposers (Rundel 1978;Chapter14). As a consequence, it is not surprising that these secondary products are frequently used by the pharmaceutical industry as anti- bacterial and antiviral compounds. In addition, lichens have long been used as a source of natural dyes and in the making of perfumes. In both cases the secondary products provide the chemical basis for these applications (Chapter 7). Introduction 7
The differential sensitivity of lichens to air pollution has been recognized for over a century and a half, and the application of lichenological studies to biomonitoring of air pollution is now well developed (Chapter 15). For example, patterns in lichen communities may be correlated with sulfur dioxide levels in the atmosphere (Hawksworth and Rose 1970). In recent years sulfur dioxide levels have been reduced, either by improved controls on emissions or by more efficient dispersion strategies, and, as a consequence, lichens are now reinvad- ing areas from which they had previously disappeared (Rose and Hawksworth 1981; Bates et al. 1990). However, the recolonization is incomplete because other factors, such as high nitrate deposition, modify lichen community composition as well. Finally, lichens are efficient accumulators of metals and persistent organic pollutants and are frequently used as surrogate receptors for document- ing deposition of these pollutants (Chapter 12).
1.6 Lichens as self-contained miniature ecosystems
The lichen thallus is a relatively stable and well-balanced symbiotic system with both heterotropic and autotrophic components. From this perspec- tive, the lichen can be regarded as a self-contained miniature ecosystem (Farrar 1976c; Seaward 1988), particularly if one considers the parasitic lichenicolous fungi colonizing lichens as this ecosystem’s decomposers. The lichen fungus undoubtedly benefits enormously by obtaining its nutrition from the photo- biont, but the photobiont’s gain from the association is less obvious. Fundamen- tally, the photobiont gains protection from high light, temperature extremes and to some extent drought, but the premise that the alliance between free- living algae or cyanobacteria and the fungal partner enables them to live together in inhospitable areas where they could not do so independently cannot be fully justified. Pushed to its ultimate limit, this train of thought leads to the fallacy that lichens are the only form of life possible on other planets – a false assumption, because, even supposing that the environment there was capable of supporting life as we know it, then representatives of both symbiotic partners would have to be present in the first instance. However, lichens have recently been put to the test in terms of their ability to cope with extreme conditions of outer space, even Martian conditions, the symbiotic system and germination capacity proving remarkably resistant to UV radiation and vacuum exposure (de Vera et al. 2003, 2004). The lichen symbiosis typically involves a close physiological integration. The usually dominant mycobiont is, of course, a heterotrophic organism that derives its carbon nutrition from the photobiont (Chapter 3). The flux of carbo- hydrates, as polyols in the case of green algal lichens and glucose in the case of 8 T. H. Nash III
cyanolichens, from the photobiont to the mycobiont is well established (Smith and Douglas 1987). This is a necessary benefit for the mycobiont and is the result of the photobiont’s cell walls being more permeable to carbohydrate loss in the lichenized than nonlichenized state (Hill 1976). In addition, the mycobiont gains a nitrogen source in the case of cyanolichens, in which nitrogen fixation occurs in the photobiont (Chapter 11). No comparable flux of nutrients from the mycobiont to the photobiont has been demonstrated. However, the recent demonstation of recycling of nitrogen and phosphorus in a mat lichen is an exciting first step to providing such documentation (Hyva¨rinen and Crittenden 2000;Elliset al. 2005). Does the fungus in general serve as a reservoir of inorganic nutrients for the photobiont through the haustoria? Certainly other fungi facil- itate nutrient uptake in other symbiotic relationships, such as occurs in mycor- rhizae and rhizospheric fungi. Another result of close physiological integration is the occurrence of a wide range of secondary products, many of which occur as crystals extracellularly within the lichens (Chapter 7). Most of these are unknown in free-living fungi (or other organisms) and hence their occurrence adds to the uniqueness of the lichen symbiosis. From an ecological perspective, lichens may be even more complex, as free- living bacteria and non symbiotic fungi may be found associated with an ‘‘indi- vidual’’ (Section 13.4) and, as a consequence, some authors regard a lichen as a miniature ecosystem. Further support for accepting the lichen as an ecosystem is provided when one considers the range of other benign or harmful micro- organisms associated with one or more of the above bionts; these include fungi and bacteria found both on the surface and within thalli, or in the microenvir- onment generated beneath thalli or within lichen-weathered substrata (Bjelland and Ekman 2005), and also invertebrates which graze upon them, or seek protection from predators through crypsis or by sheltering beneath thalli; the intimate relationship between the lichen and its substratum in the case of epiphytic, lignicolous and foliicolous species adds to the complexity of the microhabitat generated. 2 Photobionts
t. friedl and b. b del
2.1 Major differences in cyanobacteria versus algae
Nearly 40 genera of algae and cyanobacteria have been reported as photobionts in lichens (Tschermak-Woess 1988;Bu¨ del 1992). Three genera, Trebouxia, Trentepohlia, and Nostoc, are the most frequent photobionts. The genera Trebouxia and Trentepohlia are of eukaryotic nature and belong to the green algae; the genus Nostoc belongs to the oxygenic photosynthetic bacteria (cyanobac- teria). Eukaryotic photobionts are also referred to as ‘‘phycobionts’’ while cya- nobacterial photobionts are sometimes called ‘‘cyanobionts.’’ The vast majority of eukaryotic photobionts belongs to the green algae (phylum Chlorophyta) which share many cytological features and their pigmentation, e.g. the pre- sence of chlorophylls a and b, with the land plants (Bold and Wynne 1985; van den Hoek et al. 1993). Only two genera of eukaryotic photobionts containing chlorophylls a and c (phylum Heterokontophyta sensu van den Hoek et al. 1993) have thus far been reported: Heterococcus, Xanthophyceae, and Petroderma, Phaeophyceae (Tschermak-Woess 1988;Ga¨rtner 1992). Cyanobacteria are of prokaryotic nature and lack chloroplasts, mitochondria, and a nucleus, all of which are found in eukaryotic algae. In cyanobacteria, thylakoids lie free in the cytoplasm, often more or less restricted to the periph- ery. The circular DNA is not associated with histones and is concentrated in areas of the cytoplasm free of thylakoids which sometimes are called ‘‘nucleoplasm.’’ Metabolite transfer from the autotrophic photobiont to the heterotrophic mycobiont depends on the type of photobiont involved. In lichens with green algal photobionts, the carbohydrates are sugar alcohols; in lichens with cyano- bacteria it is glucose (Feige and Jensen 1992; Section 10.2.1). The mode of
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. 10 T. Friedl and B. Bu¨ del
activation of CO2 uptake is another basic feature that varies depending on whether the photobiont is prokaryotic or eukaryotic. In many green-algal lichens, positive net photosynthesis is possible after water vapor uptake alone. In contrast, in cyanobacterial lichens no measurable gas exchange occurs because the water content level required to activate photosynthesis is higher and liquid water is needed to obtain such levels (Lange et al. 1986).
2.2 Identification, reproduction, and taxonomy of photobionts
2.2.1 Cyanobacteria Identification of cyanobacterial photobionts in the intact lichen thal- lus is often impossible since the morphology of the photobiont is changed by the influence of the fungal partner. Filamentous forms may be deformed to such a degree that their originally filamentous organization cannot be recog- nized within the lichen thallus, e.g. in the genus Dichothrix (Fig. 2.1). Only the truly branched filamentous cyanobacterial genus Stigonema (Fig. 2.2) and the nonbranched genus Nostoc (Fig. 2.3) can often be identified within the lichen thallus. Furthermore, cyanobacteria do not show all characteristic stages of their life cycles in the lichenized state. Because it is essential to know these stages for positive identification of cyanobacteria, even at the genus level (Koma´rek and Anagnostidis 1998, 2005), isolation and cultivation of the cyano- bacterial photobiont are necessary steps for positive identification. The mode of vegetative cell divisions is also important in the delimitation of many uni- cellular cyanobacteria at the genus level. However, using molecular techniques, at least determination on the genus level is possible directly from the lichen thallus, using specific primers for cyanobacterial 16S rDNA (e.g. Lohtander et al. 2003; O’Brien et al. 2005). Cyanobionts with heterocysts like Nostoc (Fig. 2.3) increase heterocyst frequency up to five times when lichenized compared with the free-living state (Feige and Jensen 1992). Also, cell size of cyanobacterial photobionts may be increased compared with cultured or free-living material, as has been reported for the genera Gloeocapsa (Fig. 2.4) and Chroococcidiopsis in the lichen genera Lichinella, Peccania, Psorotichia, Synalissa, and Thyrea (Geitler 1937;Bu¨ del 1982). Increase of cell size can be a result of a very close mycobiont–photobiont contact, as in the deeply penetrating haustoria. This can be seen well in the vegetative trichome cells of Scytonema sp. within the lichen Dictyonema sericeum (Fig. 2.5). Unicellular cyanobacterial photobiont genera, e.g. Chroococcidiopsis and Myxosarcina, very rarely show their specific mode of reproduction when lichenized, but frequently show these stages when cultured. For instance, Chroococcidiopsis (Fig. 2.7) and Myxosarcina (Fig. 2.6) are characterized in culture Photobionts 11
1 2
10 µm 10 µm
3 10 µm 4
10 µm
5 6
he
ha
10 µm 10 µm
Figs. 2.1–2.6 Light microscopy of cyanobacterial photobionts. Fig. 2.1. Dichothrix sp. isolated from Placynthium nigrum, showing the characteristic branching mode of the genus. Fig. 2.2. Stigonema ocellatum, free-living sample at an early stage of licheni- zation; fungal hyphae indicated by arrows. Fig. 2.3. Nostoc sp. from Peltigera canina; 12 T. Friedl and B. Bu¨ del
by multiple fission following one or two binary divisions (Waterbury and Stanier 1978;Bu¨ del and Henssen 1983). Various cyanobacterial genera have been found forming lichens, either as the primary or secondary photosynthetic partner. Their characteristic organization is shown in Figs. 2.1–2.7. Diagnostic features as well as systematic position according to the new system suggested by Anagnostidis and Koma´rek (1985, 1988, 1990) and Koma´rek and Anagnostidis (1998, 2005) and the bacteriological approach (Castenholz and Waterbury 1989) are summarized in Table 2.1. Other genera are mentioned in the literature, but they are not included here because their identification has not yet been based on cultured material. The taxonomy of cyanobacteria found thus far in lichens is summarized in Bu¨ del (1992). The heterocyst-containing genus Nostoc (Fig. 2.3) is the most common cyanobacterial photobiont, closely followed by the unicellular genera Gloeocapsa (Fig. 2.4) and Chroococcidiopsis (Fig. 2.7). At present, taxonomy of cyanobacteria at the species level is in a state of flux. Identification of cyanobacterial isolates from lichens at the species level is, at least for many unicellular taxa, almost impossible, since the species concepts in use (e.g. Geitler 1932; Koma´rek and Anagnostidis 1998, 2005) are basically defined on ecological features. Although the close relationship of cyanobacteria (formerly called blue-green algae) and bacteria has been known for more than a century (Cohn 1853), classification of the cyanobacteria was mainly based on their morphology. As in many other groups of organisms, problems arise in applying morphological criteria to systematics because considerable variation of morphological features occurs in relation to different environmental condi- tions. Recently, however, sequence comparisons of the small ribosomal subunit RNA (16S rRNA) have led to a revised view on the systematics and phylogeny of cyanobacteria (Wilmotte and Golubic 1991; Turner et al. 2001; Fewer et al. 2002; Gugger and Hoffmann 2004; Henson et al. 2004; Svenning et al. 2005; Tomitani et al. 2006). These data support the modern concept of using such morphological criteria as mode and planes of division, cell differentiation and morphological
Caption for Figs. 2.1–2.6 (cont.) primordia of colonies with the typically apically attached primary heterocytes (arrows). Fig. 2.4. Gloeocapsa sanguinea in and at the margin of Synalissa symphorea. Fig. 2.5. Scytonema sp. in the thallus of Dictyonema sericeum; vegetative cells penetrated by mycobiont haustoria (ha), arranged in the center of cyanobiont cells along the long- itudinal axis of the filament, heterocyte (he) not penetrated. Fig. 2.6. Myxosarcina sp. isolated from Peltula euploca, just after the cyanobiont was liberated from the thallus; colonies with numerous nanocytes are surrounded by bacteria, typical for early stages of the isolation procedure. Photobionts 13
I 2
I Ca t
CW
S
1 µm
Fig. 2.7 Ultrastructure of Chroococcidiopsis sp., isolated from Psorotichia columnaris. Colony of daughter cells after two binary divisions (arrows), followed by irregular subsequent multiple fission in the upper part. S, striated sheath, CW, cell wall (outer lipoprotein bilayer, inner murein or petidoglucan layer), t, thylakoids, Ca, carboxysomes. complexity, and differences in developmental stages in the life cycle as diag- nostic features. In the 16S rRNA phylogenies, cyanobacteria with one cell type and binary division in one plane only are polyphyletic (i.e. do not have a common ancestor). The morphologically more complex groups of genera with cell division in several planes and producing nanocytes (‘‘baeocytes’’ according to Waterbury and Stanier 1978) form two distinct clusters, placing 14 T. Friedl and B. Bu¨ del
Table 2.1. Genera of cyanobacteria identified from lichens arranged according to taxonomic characters
‘‘Botanical’’ ‘‘Bacteriological’’ Taxonomical character system system
1. Unicellular 1.1 Binary division only Chroococcalesa Chroococcales Gloeocapsa, Chroococcus, Cyanosarcina, Entophysalis 1.2 Binary division þ multiple fission Chroococcalesa Pleurocapsales Nanocytes immotile Chroococcidiopsis Nanocytes motile Myxosarcina, Hyella
2. Filamentous with heterocytes 2.1 Nonbranched Nostocales Nostoc 2.2 False branching, no tapering trichomes Nostocales Scytonema 2.3 False branching, tapering trichomes Nostocales Calothrix, Dichothrix 2.4 True branching Stigonematales Stigonema, Hyphomorpha
a The order Chroococcales is subdivided into seven families in the ‘‘botanical’’ system. Source: Bu¨ del (1992).
the unicellular Chroococcidiopsis cluster as the closest living relatives to the heterocyst-forming filamentous cyanobacteria with a high statistical support. The Pleurocapsa cluster itself forms a well-supported sister group to the Chroo- coccales (Fewer et al. 2002). The 16S rDNA data confirm that heterocyst-forming cyanobacteria are monophyletic. Within that cluster, the Nostocales are divided into three groups (Fig. 2.8). One is well supported and includes the filamentous, nonbranched, and heterocyst-containing taxa (e.g. Nostoc). The second, also well supported, cluster includes the filamentous, heterocyst-containing taxa with false branching (e.g. Scytonema). The third cluster is not supported and contains filamentous, heterocyst-forming taxa with false branching and/or tapering tri- chomes (e.g. Tolypothrix, Calothrix). The morphologically most complex group of cyanobacteria, the former Stigonematales, with true branching, heterocyst- forming trichomes (e.g. Fischerella, Hapalosiphon, and Stigonema) have low statis- tical support and are polyphyletic. This has been conclusively demonstrated earlier (Gugger and Hoffmann 2004). Photobionts 15
Pleurocapsa sp., PCC 7516 (X78681) 100 /100
98 /100 Myxosarcina sp., PCC 7312 (AJ344561)
Xenococcus sp., PCC 7305 (AF132783) Pleu
98 /100 Gloeocapsa sp., PCC 73106 (AB039000) 100 /100
Gloeocapsa sp., PCC 73106 (AF132784)
–/– PCC 6501 (X8680) 99 /100 Gloeothece membranacea,
Cyanothece sp., PCC 7424 (AF132932) Chro
Chroococcidiopsis sp., BB82.3 (AJ344553) Anema nummularium 97 /100
99 /100 Chroococcidiopsis sp., BB96.1 (AJ344555) Peltula euploca 67 / 79 Chroococcidiopsis sp., BB79.2 (AJ344552) Thyrea pulvinata Pleu
Nostoc sp., Türk 19926 (DQ185233) Lobaria amplissima
97 / 100 Nostoc punctiforme, SAG 71.79 (DQ185258)
100 /100 Nostoc sp., O’Brien 01081201 (DQ185219) Peltigera rufescens – / 78
Nostoc muscorum, SAG 57.79 (DQ185254) – / 98 55/100 Nostoc sp., O’Brien 020708-0-9-1 (DQ185203) Peltigera membranacea 95 /100 95 /100 Nostoc commune, O’Brien 0201 1101 (DQ185223) 54 / 98 Nostoc sp., O’Brien 01072402 (DQ185216) Collema crispum
100 /100 62 / 85 Nostoc sp., PCC 6720 (DQ185240) Nost
97 / 100 Fischerella muscicola, PCC 7414 (AB039003)
– / 77 Hapalosiphon welwitschii, (AY034793)
– / 85 Stigonema ocellatum, SAG 48.90 (AJ544082) Stig
Calothrix desertica, PCC 7102 (AF132779)
–/– 80 /100 Tolypothrix sp., CCMP1185 (AB075998) Nost
PCC 71 10 (AF132781) 99 /100 Scytonema hofmannii,
Scytonema sp., U-3-3 (A Y069954) Nost
Lyngbya aestuarii, PCC 7419 (AB039013) 98 / 60
Oscillatoria sancta, PCC 7515 (AB039015) Osci
Gloeobacter violaceus, PCC 7421 (AF132790) 0.01 substitutions/site
Fig. 2.8 16S rDNA phylogeny of cyanobacterial genera that also occur as cyano- bionts (F. Kauff and B. Bu¨ del, unpublished data). Gloeobacter violaceus was added as outgroup. Chro, Chroococcales; Nost, Nostocales; Osci, Oscillatoriales; Pleu, Pleurocapsales; Stig, Stigonematales. Bootstrap frequencies (number in italics) and posterior probabilities are indicated above horizontal branches. Phylogram gener- ated with RAxML-HPC-2.1.2 Stamatakis (2006) out of 200 replicates and a GTRMIX model. Bootstrap proportions calculated with 500 replicates. Posterior probabilities estimated with MrBayes3.1.1 Huelsenbeck and Ronquist (2001), generating 20 000 000 generations using a GTR model with gamma distribution and a proportion of invariable sites. GeneBank accession number given in brackets. 16 T. Friedl and B. Bu¨ del
2.2.2 Green algae The organization of green-algal photobionts is simple: only coccoid (Figs. 2.9–2.15), sarcinoid or filamentous (Figs. 2.16–2.21) forms are known. No flagellates are known from lichens. Furthermore, filamentous forms are often reduced to short filaments (Fig. 2.20) or even to unicells (Fig. 2.17) within
9 10
m
p
m
10 µm 10 µm
11 12
m
n
p
10 µm 10 µm
13 14 15 10 µm 10 µm c
at
c n sp n es
10 µm ve
Figs. 2.9–2.15 Trebouxia spp. and other coccoid species, lichenized and in culture. Fig. 2.9. Trebouxia gigantea within a thallus of Xanthomaculina hottentotta. Unicellular stages attached to mycobiont hyphae (m). Note prominent pyrenoid (p) in the center of the algal chloroplast. Fig. 2.10. Trebouxia gigantea in culture, isolated from Photobionts 17 lichen thalli. In lichens, green algae, except those belonging to the order Trentepohliales, reproduce only asexually by sporulation (i.e. forming motile zoospores with flagella and/or immotile autospores which lack emergent fla- gella) or true cell division (for terminology see Sluiman et al. 1989). If flagellated stages (zoospores) can be formed (e.g. in Trebouxia), they are released frequently in culture, but in the lichen thallus they may be observed only occasionally (Slocum et al. 1980). Identification of green lichen algae at the genus level is often possible without culturing, e.g. by simple squash preparations of the algal layer. For identification of the species, cultures are essential because important features such as chloroplast morphology or certain stages of the life cycle may be reduced or absent in the lichenized state. Most green lichen algae are only facultative photobionts, i.e. they are also occur independently as epiphytes, endoliths or as soil algae. In modern taxonomic concepts of the Chlorophyta (e.g. Mattox and Stewart 1984; van den Hoek et al. 1993) the classes of green algae are regarded as different evolutionary lineages with each identified by a unique type of flagel- lated stage (zoospores or gametes). In contrast to traditional taxonomies, the former single class Chlorophyceae is now split into several separated new classes. The class Pleurastrophyceae with the order Pleurastrales (Mattox and Stewart 1984) and, independently, the new order Microthamniales (Melkonian 1990), have been established to classify the most common green lichen photo- biont Trebouxia and some other morphologically diverse green algae (e.g. Pleurastrum terrestre, Microthamnion kuetzingianum) which share, however, unique
Caption for Figs. 2.9–2.15 (cont.) Xanthomaculina hottentotta. Autospore packages are dominant in the culture four weeks after the alga was liberated from the mycobiont. Fig. 2.11. Trebouxia arboricola within a thallus of Omphalora arizonica, m, mycobiont hypha. Pyrenoid (p) in the center of the chloroplast, nucleus (n) located externally in an invagination of the chloroplast. Fig. 2.12. Trebouxia gelatinosa, vegetative cells in culture. Note crinkled chloroplast. Fig. 2.13. Asterochloris sp. isolated from Diploschistes albescens in culture. Vegetative cells (ve) with deeply incised and crinkled chloroplast, nucleus (n) located externally in an invagination of the chloroplast. Smooth chloroplast appressed to cell wall in early stage of sporangium development (sp). Sporangia with typical cap-like wall thickening (c). Fig. 2.14. Vegetative cells of Coccomyxa subellipsoidea isolated from Botrydina vulgaris in culture (strain SAG 216–13). Vegetative cells elongated and of irregular shape with a flat chloroplast appressed to the cell wall and without pyrenoids. (es) empty walls of autosporangia. Photograph taken by I. Kostikov. Fig. 2.15. Vegetative cells of Myrmecia biatorellae isolated from Lobaria linita in culture (strain SAG 8.82). Chloroplast smooth and attached to the cell wall, nucleus (n) with nucleolus located centrally; at, autospore package. Photograph taken by T. Darienko. 18 T. Friedl and B. Bu¨ del
16 17 18 sp v
sp
n at as n at 10 µm
10 µm 10 µm as
19 20 21
m
d
20 µm 10 µm 20 µm
Figs. 2.16–2.21 Fig 2.16. Dictyochloropsis reticulata isolated from Brigantiaea ferruginea in culture (strain SAG 2150). Reticulate chloroplast attached to the cell wall, nucleus (n) located in the center of the cells. at, autospore package. Fig. 2.17. Elliptochloris bilobata in culture, isolated from subaerial habitats (strain SAG 245.80). Flat chloroplast appressed to cell wall, center of the cell with nucleus (n) and several large vacuoles (v). Sporangia at different developmental stages (sp). Note autospores (as) of different shape which is characteristic for the genus, elongated (right) and more spherical (middle). Fig. 2.18. Filamentous stages of Leptosira obovata in culture, isolated from freshwater (strain SAG 445–1). at, autosporangium. Fig. 2.19. Branched filament of Trentepohlia sp., free- living sample. Cells filled with carotenoid droplets covering the chloroplasts. Note lamellate cell wall (arrow). Fig. 2.20. Short and unbranched filament of a lichenized Trentepohlia sp. within a thallus of Enterographa subpallidella. Note carotenoid droplets (d) and mycobiont particle (m). Fig. 2.21. Young branched filament of Dilabifilium arthropyreniae in culture, isolated from Arthropyrenia kelpii (strain SAG 467–2).
ultrastructural features of their zoospores and their mitosis/cytokinesis pat- terns (Melkonian and Peveling 1988; Melkonian 1990). The taxonomic position of the order Microthamniales, however, is uncertain. It is regarded as a distinct lineage of green algae closely related to the class Chlorophyceae s. str. (Melkonian 1990) or as an order of the class Ulvophyceae (Sluiman 1989). Most recent approaches to green-algal systematics also include molecular data (Figs. 2.24, 2.25). Gene sequence comparisons of the small ribosomal Photobionts 19
Fig. 2.22 Schematic drawings of some chloroplast types in Trebouxia. Left part of cells, surface view; right part of cells, optical section. (a) T. crenulata, (b) T. corticola, (c) T. jamesii, (d) T. irregularis (from S. Takeshita, unpublished data). subunit RNA (18S rRNA) support the ultrastructural findings that Trebouxia forms together with Pleurastrum terrestre and Microthamnion kuetzingianum a dis- tinct group of green algae, the Microthamniales (Friedl and Zeltner 1994). This order is evolutionarily distinct from the Ulvophyceae, but shares a sister group relationship with the Chlorophyceae s. str., while the Pleurastrophyceae are polyphyletic in their rRNA phylogenies (Steinko¨ tter et al. 1994). Based on these data, Trebouxia forms together with other coccoid lichen and soil algae (Myrmecia biatorellae, Friedmannia israelensis) an independent lineage, the ‘‘Lichen Algae Group’’ within the Microthamniales (Friedl and Zeltner 1994). A great variety of morphologically diverse green algae has been identified from lichens. A complete list of all green-algal taxa found as photobionts so far was presented by Tschermak-Woess (1988). Examples of green-algal photobionts are given below. Trebouxia spp. (Figs. 2.9–2.12, 2.22, 2.23) are the most common green photo- bionts. Within lichen thalli, only nonmotile stages with a reduced chloroplast occur (Figs. 2.9, 2.11). In culture, different patterns of autospore formation (e.g. autospore packages present [Fig. 2.10], cell cycle A; or autospore packages absent, cell cycle B; Friedl 1993) and motile stages (zoospores) are formed for reproduc- tion. The nucleus is located in an invagination of the chloroplast (Figs. 2.11, 2.12). Vegetative cells exhibit a central massive or star-shaped chloroplast that is 20 T. Friedl and B. Bu¨ del
1 µm
Fig. 2.23 Electron microscopy of a pyrenoid of Trebouxia impressa, lichenized phycobiont within a thallus of Parmelia sulcata.
wrinkled in different ways (Figs. 2.11, 2.12, 2.22) and contains several pyrenoids with a diverse ultrastructure (Fig. 2.23; Friedl 1989a). Several distinct patterns of chloroplast lobes are formed (Ettl and Ga¨rtner 1984), and these are used to identify species of Trebouxia (Ga¨rtner 1985). Twenty-five species are recognized by Ga¨rtner (1985), and 16 species by Friedl (1989b). Trebouxia was formerly split into two genera, Trebouxia and Pseudotrebouxia, based on differences seen in the mode of reproduction. Cell packages that occur only in some species (‘‘Pseudotre- bouxia spp.’’) but not in others (‘‘Trebouxia spp.’’) were believed to be the result of true cell division and therefore fundamentally different from autospores and zoospores (Archibald 1975; Hildreth and Ahmadjian 1981). However, this separa- tion has not been supported by recent studies of autospore formation patterns (Ettl and Ga¨rtner 1984; Friedl 1993), chloroplast characters (Ga¨rtner 1985; Friedl 1989a), and zoospore ultrastructure (Melkonian and Peveling 1988). Gene sequence comparisons of the large ribosomal subunit (25S rDNA) support chlor- oplast morphology (Ettl and Ga¨rtner 1984) as an important character for the distinction of Trebouxia species (T. Friedl and C. Rokitta, unpublished). In molecu- lar phylogenies based on gene sequences of the small and large ribosomal sub- units, Trebouxia appears as a paraphyletic genus (Friedl and Zeltner 1994; T. Friedl and C. Rokitta, unpublished). This supports the division of Trebouxia into at least two genera, but these findings do not coincide with the concept of Pseudotrebouxia sensu Archibald (1975). It is not yet clear whether Trebouxia is an obligate symbiotic genus of green algae or whether it also occurs free-living (Ahmadjian 1988, 1993). Photobionts 21
Trebouxiophyceae Trebouxia arboricola SAG 219-1a Z68705 Trebouxia asymmetrica SAG 48.88 Z21553 Trebouxia impressa UTEX 892 Z21551 Trebouxiales Myrmecia astigmatica ASIB T76 Z47208 Myrmecia biatorellae UTEX 907 Z28971 Asterochloris phycobiontica SAG 26.81 “Trebouxia” magna UTEX 902 Z21552 Coelochlamys perforata UTEX 2104 M62999 Microthamnion kuetzingianum UTEX 1914 Z28974 Leptosira terrestris SAG463-3 Z28973 Coccobotrys verrucariae SAG 16.97 Desmococcus olivaceus SAG SAG 1.92 Stichococcus bacillaris SAG 397-1b AJ416107 Prasiola crispa SAG 43.96 AJ416106 Diplosphaera mucosa SAG 48.86 Prasiolales “Chlorella” sphaerica SAG 11.88 Aj416105 “Chlorella” mirabilis Andreyeva 748-I X74000 “Chlorella” ellipsoidea SAG 211-1a X63520 Coccomyxa peltigerae SAG 216-6 Elliptochloris bilobata SAG 245.80 Botryococcus braunii AJ581912 Choricystis minor SAG 251-1 X89012 Dictyochloropsis symbiontica SAG 27.81 Dictyochloropsis reticulata SAG 2150 Z47207 “Chlorella” trebouxioides SAG 3.95 “Chlorella” saccharophila SAG 211-9a X63505 Micractinium pusillum SAG 7.93 AF499921 Parachlorella kessleri SAG 211-11g X56105 Chlorellales Chlorella vulgaris SAG 211-11b X13688
Chlorophyceae Ulvophyceae
Dilabifilum arthropyreniae SAG46.72 Ulva intestinalis AF189077 Urospora penicilliformis AB049417 Ulothrix zonata Z47999 Trentepohlia aurea DQ399590 Trentepohlia sp. UTEX1227 Cephaleuros virescens SAG 28.93 Oltmannsiellopsis viridis NIES 360 D86495 AY220984
Dangemannia microcystis SAG 2022 AJ416104
Scherffelia dubia SAG 17.86 X68484 CHLOROPHYTA prasinophytes Tetraselmis striata Ply 443 X70802 Nephroselmis olivacea SAG 40.89 X74754 STREPTOPHYTA prasinophytes Nephroselmis pryriformis CCMP 717 X75565 Charophytes and Embryophytes
Mesostigma viride SAG50-1 AJ250108 prasinophyte
0.02
Fig. 2.24 Phylogeny of green-algal photobionts and nonlichenized green algae inferred from 18S rDNA sequence analyses. Species representing lichen photobionts are marked with a filled bar. A circle indicates a lineage of green algae in which lichen symbionts have evolved. Lichen photobionts are closely related to nonlichenized 22 T. Friedl and B. Bu¨ del
The small colonies that have been observed (e.g. Tschermak-Woess 1978)onbark and soil a few times could also have escaped from damaged lichen thalli. Myrmecia biatorellae differs from Trebouxia spp. by the cup-shaped chloroplast that lacks pyrenoids and occupies a parietal position, and the nucleus is located centrally (Fig. 2.15). Species of Dictyochloropsis are characterized by their reticu- late chloroplast (Fig. 2.16). A close relationship of Myrmecia biatorellae with Trebouxia spp. is suggested by ultrastructural (Deason 1989) and molecular data (Friedl and Zeltner 1994) while the taxonomic position of Dictyochloropsis spp. is still unclear. Species of Myrmecia and Dictyochloropsis are also found as soil algae and on bark (Nakano et al. 1991). In several other coccoid green photobionts, flagellated stages are unknown even in culture and reproduction is performed by autospores exclusively. Thus, a proper classification of these autosporic taxa remains unclear. Autosporic phycobionts include species of Chlorella and several other Chlorella-like algae, e.g. Coccomyxa (Fig. 2.14), Elliptochloris, Diplosphaera and Nannochloris (Tschermak- Woess 1988). Cells of Chlorella and Chlorella-like algae have a spherical or ellipsoi- dal shape, are often minute in size (some are less than 10 mm in diameter), contain a simple parietal chloroplast (Fig. 2.14) and some produce a gelatinous matrix (e.g. Coccomyxa). Due to their small size and lack of easily studied char- acters, taxomomy of these algae is only poorly understood. Differences seen in the formation of autospores may be a distinctive feature of important taxonomic value (Ga¨rtner 1992). Chlorella is polyphyletic (Huss and Sogin 1990). Diplosphaera and Nannochloris may be closely related to each other (Ga¨rtner 1992). Chlorella spp.
Caption for Fig. 2.24 (cont.) species. Note that there are several independent origins for the symbiotic lifestyle within the classes Trebouxiophyceae and Ulvophyceae. The phylogeny shows the deep division of green algae into two clades, Chlorophyta and Streptophyta. The green-algal class Chlorophyceae and the group comprising charophytes and embry- ophytes which do not contain lichen photobionts are shown as diamonds, which stand for sequences not shown in the graphic but are used for the calculation of the tree. The code next to a species name is the accession number for a culture strain when available from a public culture collection. The other code refers to the GenBank accession number of the sequence; sequences without GenBank accession numbers are still unpublished. A distance phylogeny (neighbor-joining method; Hasegawa et al. [1985] model) is shown, on which statistical support ( > 70% in bootstrap tests, > 0.7 a priori probabilities) using a maximum-likelihood model in conjunction with a mini- mum evolution distance approach, maximum parsimony and maximum likelihood (Bayesian inference; Huelsenbeck and Ronquist 2001) is indicated by thick internal branches. The phylogeny was rooted by two species of Glaucophyta which have been pruned away from the graphic. Photobionts 23
T. arboricola SAG 219-1a Z68705
T. crenulata CCAP 219/2
T. decolorans UTEX 901
T. incrustata UTEX 784 AJ293795
arboricola T. gigantea UTEX 2231 AJ249577 – Clade T. showmanii UTEX 2234 AF242470
T. asymmetrica SAG 48.88 AJ249565 T. jamesii SAG 2103 impressa/gelatinosa T. gelatinosa UTEX 905 Z68698 – Clade T. anticipata UTEX 903
T. flava UTEX 181 AF242467
T. impressa UTEX 892 AF345891 simplex – Clade T. potteri UTEX 900 AF242469
T. “lethariae” AF242460
T. simplex SAG 101.80
T. “hypogymniae” AJ511358
T. “angustilobata” SAG 2204 AF128271
T. usneae UTEX 2235 AJ249573 corticola – Clade T. corticola UTEX 909 AJ249566
T. higginsiae UTEX 2232 AJ249574
T. galapagensis UTEX 2230 AJ249567
0.05 subst./site
Fig. 2.25 Phylogeny of species of the green-algal photobiont Trebouxia from culture inferred from ITS rDNA sequence analyses. Four deeply diverging clades of species (marked by circles on internal branches) are resolved and are named according to a certain species from the clade. Schematic drawings for chloroplast morphology as viewed by light microscopy (left) and pyrenoid ultrastructure (right; Friedl 1989a) are shown next to each clade; these features define clades within Trebouxia. Only sequences from authentic cultures were used for this phylogeny. Authentic cultures are valuable as references; they represent the culture material on which the taxonomic description of a species was based. Species names in quotation marks indicate that a formal taxonomic description for this species is still pending. The code given next to a species name represents the accession numbers for a culture strain if available from public culture collections. The other code refers to the GenBank accession number of the sequence if available from public sequence data bases. Sequences without GenBank accession numbers are still unpublished. An unrooted maximum likelihood phylogeny is shown on which statistical support ( > 70% in bootstrap tests, > 0.7 a priori probabilities) using a maximum-likelihood model in conjunction with a minimum evolution distance approach, maximum parsimony and maximum likelihood (Bayesian inference; Huelsenbeck and Ronquist 2001) is indicated by thick internal branches. 24 T. Friedl and B. Bu¨ del
and Chlorella-like algae are well known from diverse aquatic and aerophytic habitats; Chlorella spp. are also common as endosymbionts of invertebrates. Coccobotrys and Desmococcus have sarcinoid growth habits or form multiseriate filaments (Zeitler 1954). Both genera are also known from endolithic habitats (Broady and Ingerfeld 1993). Taxonomically these genera have not yet been studied adequately. One of the most common green photobionts is the filamentous genus Trentepohlia (Figs. 2.19, 2.20), which also grows as an epiphyte (e.g. on moist rocks or bark, even on tree leaves in the tropics). In these habitats, free-living Trentepohlia forms branched filaments of cylindrical cells with thick lamellate cell walls (Fig. 2.19) and several parietal chloroplasts. Zoospores and gametes are produced in specialized cells differing from vegetative cells. When grown as an epiphyte, the protoplast can be entirely filled by droplets of carotenoids (Fig. 2.19) and then Trentepohlia forms orange or reddish plant masses. The same pigmentation is present in lichenized Trentepohlia (Fig. 2.20), giving rise to an orange color of the algal layer. So, scratching a lichen thallus surface causes an orange appearance, and one can safely predict that Trentepohlia is the photo- biont. Within lichen thalli, Trentepohlia forms only short and thin filaments (Fig. 2.20) or consists of unicellular stages. Taxonomic relationships of the Trentepohliales are uncertain; some ultrastructural zoospore characters are shared with the Ulvophyceae (Chapman 1984). Other filamentous green photobionts are Pleurastrum terrestre and species of Dilabifilium (Fig. 2.21). Pleurastrum terrestre forms short uniseriate filaments of elongated cells with a cup-shaped chloroplast in liquid culture media, but is unicellular within lichen thalli. Pleurastrum terrestre is known as a photobiont only from the lichen genera Vezdaea and Thrombium (Tschermak-Woess and Poelt 1976; Tschermak-Woess 1988), but has often been isolated from soil samples (Tupa 1974). Dilabifilium (including Pseudopleurococcus) forms branched filaments of cylindrical cells and has been found in primitively organized crustose aquatic lichens (e.g. Verrucaria and Arthropyrenia; Tschermak-Woess 1976), but is also known as an epiphytic alga in marine habitats and freshwater (Johnson and John 1990; Ihda et al. 1993). Pleurastrum terrestre is a member of the Microthamniales and is synonymous with Leptosira obovata. The taxonomic position of the genus Dilabifilium is still unclear (Johnson and John 1990).
2.3 Occurrence within lichens
Most lichen species contain green algae as photobionts. Among the lichenized families of the Lecanorales (nomenclature of lichen orders according to Henssen and Jahns 1974), Trebouxia is the most frequent photobiont, while Photobionts 25
Trentepohlia is more frequent in lichen genera of the Arthoniales (e.g. Roccella), Gyalectales (e.g. Coenogonium) and the Sphaeriales. Members of the Trentepoh- liales are also common green algal photobionts of epiphyllic lichens in the tropics. In the Ostropales (e.g. Graphis, Diploschistes) both Trebouxia and Trentepoh- lia are equally frequent. Chlorella and Chlorella-like algae are most frequent in the Caliciales, but also occur in some crustose lichens of the Lecanorales (e.g. Lecidella, Micarea, Trapelia). Coccomyxa is common in the families Baeomyce- taceae and Peltigeraceae as well as in lichenized Basidiomycetes. In the Peltiger- aceae and the Stictaceae, Dictyochloropsis is the most common green algal photobiont. Myrmecia is a common photobiont in the genus Dermatocarpon (including Catapyrenium, Verrucariales). The order Verrucariales is rather diverse with respect to their photobionts, as the green algae Coccobotrys, Desmococcus and Dilabifilium as well as the heterokonts Heterococcus (Xanthophyceae) and Petroderma (Phaeophyceae) have all been found as photobionts in species of Verrucaria species. Furthermore, the same lichen species can contain different species of Trebouxia, i.e. one mycobiont can form morphologically identical thalli with different algal species. For example, three different species of Trebouxia (T. arboricola, T. irregularis, and T. jamesii) have been isolated from Parmelia saxatilis (Friedl 1989b). Other examples are known from species of Anzia (Ihda et al. 1993) and Diploschistes (Friedl and Ga¨rtner 1988). Only about 10% of all lichen species contain a cyanobacterium as the primary photobiont. The Collemataceae, Heppiaceae, Lichinaceae, Peltulaceae, and Placynthiaceae have cyanobacteria as the only photosynthetic partner. In con- trast, lichens of the families Arthopyreniaceae, Coccocarpiaceae, Corticiaceae, Pannariaceae, Peltigeraceae, and Stictaceae may have either green algal or cyanobacterial photobionts. Some lichen genera have a green alga as the primary photobiont and a cyanobacterium as a secondary one. In such cases the cyanobacterium is located either in external or internal cephalodia (Chapter 5). In addition to heterocyst- producing filamentous cyanobacteria, a number of strains of the genus
Chroococcidiopsis are also capable of N2-fixation under microaerobic or anaerobic conditions (Stewart 1980; Boison et al. 2004).
2.4 Isolation and maintenance of cyanobionts and phycobionts
Techniques for the isolation of photobionts have been described sev- eral times, and the interested reader is referred to special publications like the ‘‘Handbook of Lichenology’’ (Galun 1988a) or Ahmadjian (1967b) for reference. Here, we only can give some general instructions for the isolation procedures. 26 T. Friedl and B. Bu¨ del
2.4.1 Cyanobionts From a washed thallus, fragments of the cyanobacterial layer are trans- ferred under sterile conditions to agar plates containing a mineral medium. The agar plates are then kept under low light intensities (10–30 mmol m 2 s 1 PPFD) and moderate temperatures (15–25 8C). After a few days or weeks, cyanobiont cells start to develop free from the mycobiont and then can be transferred to fresh agar plates under a dissecting microscope using a micropipette or a needle. For long-term cultivation, strains of isolated cyanobionts are kept on agar slants at low light intensities and temperatures at about 15–17 8C to keep growth at slow rates.
2.4.2 Phycobionts Isolation of green-algal photobionts is similar to that of cyanobacteria. From washed lichen thalli, a squash preparation of the algal layer is made. From this suspension of algal cells and mycobiont fragments, either single algal cells are isolated under microscopical observation using micropipettes or the whole suspension is transferred onto agar plates. The ‘‘micropipette method’’ (Ware´n 1918–19) usually results in clonal and axenic cultures. The ‘‘whole suspension’’ method is easier and the chances that the algae will grow in culture are better, but the resulting algal colonies need further purification. Most green-algal photobionts grow easily in culture. They are best maintained on liquid or agarized mineral media (Bischoff and Bold 1963; Friedl 1989a). In comparison with their growth on a mineral medium alone, some green-algal photobionts (e.g. Trebouxia) grow much faster and in larger quantities after the addition of glucose and proteose peptone to the culture medium and consequently are considered facultative heterotrophs (Ahmadjian 1967b). Most green-algal photo- biont cultures require low light intensities and moderate temperatures (about 10–30 mmol m 2 s 1 PPFD and 15 8C). The following culture collections maintain a large variety of lichen photo- bionts: the ‘‘Culture Collection of Algae at the University of Texas at Austin (UTEX)’’, Austin, Texas, USA, the ‘‘American Type Culture Collection (ATCC)’’, Rockville, Maryland, USA (cyanobacterial photobionts only), the ‘‘Sammlung fu¨r Algenkulturen (SAG)’’, Go¨ ttingen, Germany, and the ‘‘Pasteur Culture Collection (PCC)’’, Paris, France (cyanobacterial photobionts only). Additional culture strains are available from other culture collections which are listed in Myachi et al.(1989). 3 Mycobionts
r. honegger
Lichen-forming fungi (also termed lichen mycobionts) are, like plant pathogens or mycorrhizal fungi, a polyphyletic, taxonomically heterogeneous group of nutritional specialists (Tables 3.1 and 3.2) but otherwise normal repre- sentatives of their fungal classes. Long after the discovery of the dual nature of lichens by Schwendener (1867; Honegger 2000) and his proposal to include lichens in fungi, most biologists and even the majority of lichenologists con- sidered lichens as a group of organisms that differ so fundamentally from all others that they had to be treated as a separate group, e.g. as a phylum ‘‘Lichenes’’; this term is nowadays obsolete. Even in the early twenty-first cen- tury, many scientists consider lichens as plants, thus ignoring the fact that species names of lichens refer to the fungal partner, fungi forming a separate kingdom. It is the heterotrophic mycobiont of morphologically advanced lichens that mimics plant-like structures. In this chapter the similarities and differences between lichen-forming and nonlichenized fungi are discussed at the phylogenetic, morphological and cytological levels and also with regard to different nutritional strategies.
3.1 Lichenized versus nonlichenized fungi
3.1.1 Lichenization: a successful nutritional strategy Fungi, as heterotrophic organisms, have developed various nutritional strategies for acquiring fixed carbon (Table 3.1). Lichenization, i.e. the acquisi- tion of fixed carbon from a population of minute, living algal and/or cyanobac- terial cells, is a common and widespread mode of nutrition. One out of five fungal species is lichenized (Table 3.1). Some lichen-forming fungi belong to
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. 28 R. Honegger
Table 3.1. Acquisition of fixed carbon by Fungi and fungus-like Protoctista
Degradation of dead organic matter Saprobes c. 45–50%
Symbioses with C-autotrophs or C-heterotrophs: Parasitic symbioses (biotrophic or necrotrophic) with Cyanobacteria Algae Plants Fungi (lichenized and nonlichenized) c. 20% Animals Humans Mutualistic symbioses Mycorrhizae (c. 8%) Lichens (c. 21%) c. 30% Mycetocytes, etc.
Sources: Lewis (1973); Hawksworth et al.(1983).
Table 3.2. Orders of Ascomycota and Basidiomycota, which include lichenized taxa
Phylum Subphylum Class Nutritional strategies Subclass (predominant form in Thallus anatomy in Order bold characters) lichenized taxa9
Ascomycota1,2,3 Pezizomycotina Arthoniomycetes Arthoniales l, nl (sap, lp) ns,s Dothideomycetes Patellariales nl (sap, lp), l ns Eurotiomycetes Pyrenulales nl (sap), l ns Verrucariales l, nl (sap, lp) ns,s Mycocaliciales nl (sap), l Lecanoromycetes Acarosporomycetidae Acarosporales lns Ostropomycetidae Agyriales l, nl (sap) ns Gyalectales lns Mycobionts 29
Table 3.2. (cont.)
Ostropales nl (sap, pp, lp), l ns Pertusariales lns Trichotheliales lns Lecanoromycetidae Lecanorales8 l, nl (sap, lp) ns, s Peltigerales ls,ns Teloschistales l, nl (lp) ns, s Lichinomycetes Lichinales lns,s
Basidiomycota4,5 Hymenomycetes Agaricales nl (sap, myc, pp, lp), l ns,s Polyporales nl (sap, myc, pp, f, lp), l ns,s
Anamorphic fungi6 nl,l ns
Sterile taxa with no known reproductive structures7 ns
Abbreviations: f, fungicolous; l, lichenized; lp, lichenicolous (lichen parasite); myc, mycorrhiza; nl, nonlichenized; ns, nonstratified; pp, plant pathogens; s, internally stratified; sap, saprotrophic. 1 c. 98% of lichen-forming fungi are Ascomycetes. 2 c. 42% of Ascomycetes are lichenized (c. 13 500 spp.), all belonging to subphylum Pezizomycotina. 3 15 out of 52 orders of Pezizomycotina include lichenized taxa, 5 of them being exclusively lichenized. c. 0.4% of lichen-forming fungi are Basidiomycetes. 5 Only c. 0.3% of Basidiomycetes are lichenized (c. 50 spp.). 6 c. 1.5% of lichen-forming fungi (c. 200 spp.) belong to the Anamorphic fungi. 7 c.75 species of lichen-forming fungi are sterile (disperse via thallus fragmentation). With molecular data sets the taxonomic affiliation of some of these taxa will be identified. 8 Lecanorales are the largest order, comprising c. 5500 species, >99% of them lichenized. Sources: Hawksworth (1988a); Honegger (1992); Hawksworth and Honegger (1994); Kirk et al. (2001); Eriksson (2006b).
The majority of lichen-forming fungi ( >55%) form nonstratified (crustose, microfilamentous, etc.) thalli, c. 20% form squamulose or placodioid thalli, and c. 25% form morphologically advanced, foliose or fruticose thalli with internal stratification. orders with uniform nutritional strategies; others belong to orders with diverse strategies (Table 3.2). A high percentage of lichen-forming fungi are ecologically obligate, but physiologically facultative biotrophs (organisms that obtain nutrients from a living host). In other words they can be cultured in the aposymbiotic 30 R. Honegger
(‘‘free-living’’) state but in nature almost exclusively the symbiotic phenotype is found. Nonlichenized germ tubes or other free hyphae of lichen mycobionts certainly exist in natural ecosystems, but, due to their notoriously slow growth rates, they cannot be recovered with conventional isolation techniques. Molecular phylogenies elucidate taxonomic relationships among lichenized and nonlichenized fungi. Until recently lichenization was thought to be an ultimate state. Among the most fascinating mycological discoveries of recent years is the finding that lichenization can be transient. Based on thorough analyses of large sets of molecular data from lichenized and nonlichenized ascomycetes using a Bayesian phylogenetic tree sampling methodology, com- bined with a statistical model of trait evolution, Lutzoni et al.(2001)demonstrated that (1) lichens evolved earlier than previously assumed, (2) gains of lichenization were distinctly less frequent during ascomycete evolution than previously assumed, and (3) lichen symbiosis was lost several times. Consequently, numer- ous taxa of nonlichenized ascomycetes, such as Eurotiomycetidae, derive from lichen-forming ancestors. This particular group comprises economically impor- tant taxa such as the genera Penicillium and Aspergillus, with numerous species used in biotechnology because of their interesting secondary metabolism, a trait shared with many of their ancestors. The ‘‘fungal branch of the tree of life’’ is currently under construction, and many new insights in phylogenetic relation- ships of lichen-forming and nonlichenized fungi are expected in the near future (Lutzoni et al. 2004).
3.1.2 Fossil records Fossil records of lichens are extremely rare. Perhaps palaeontologists have not yet adapted their eyes to recognizing lichen-forming fungi and their photobionts. Two recent discoveries of lichen-like fossils support the view that lichenization might be a very ancient nutritional strategy. In marine phosphor- ites of the Doushantuo Formation (approx. 600 million years before present [MaBP]) from South China, and in the famous Early Devonian Rhynie chert beds in Scotland (approx. 480 MaBP), colonies of coccoid cyanobacteria or unicellular algae with mucilaginous extracellular sheaths were found, which are invaded by fungal hyphae (Taylor et al. 1995b, 1997; Yuan et al. 2005). This situation resembles the mycobiont–photobiont interface in various genera of Lichinaceae (Henssen 1963, 1986;Bu¨del1987;Henssen1995), all with cyanobac- terial photobionts (examples in Honegger 2001), or in Epigloea spp. (ascomycetes incertae sedis), which are symbiotic with or at least live within the mucilaginous colonies of Coccomyxa dispar, a unicellular green alga forming massive gelatinous sheaths (Jaag and Thomas 1934;Do¨ bbeler 1984; David 1987). Winfrenatia reticulata, the Early Devonian fossil, was assumed to be formed by a Mycobionts 31 zygomycete (Taylor et al. 1995a, b, 1997). As extant zygomycetes are not forming lichen symbioses, some investigators hesitate to interpret Winfrenatia as a lichen. However, lichenization can be lost in the course of time, as shown in ascomycete evolution (Lutzoni et al. 2001, 2004). Among the Early Devonian Rhynie chert bed fossils are beautifully preserved arbuscular mycorrhizae (Taylor et al. 1995a), strikingly similar to the ones which are nowadays symbiotic with more than 70% of higher plants. These arbuscular mycorrhizae represent a fungal symbiosis that had already reached an astonishing level of morphologi- cal and physiological complexity, although the early vascular plants (Rhyniales) were just starting to colonize terrestrial ecosystems. Terrestrial soil and rock surfaces had certainly been colonized by cyanobacteria and green algae long before the advent of vascular plants. With high probability groups of fungi have formed manifold interactions with these early photoautotrophic inhabitants of terrestrial ecosystems, ranging from parasitism to mutualism. The Protoli- chenes hypothesis of Eriksson (2005, 2006a) proposes the origin of the sub- phylum Pezizomycotina among nutritional specialists symbiotic with algae and/or cyanobacteria, so-called Protolichenes, whence extant lichenized and nonlichenized groups evolved. Fossil arbuscular mycorrhizae show that complex fungal interactions with photoautotrophic partners were already differentiated 480 Ma ago. Presently known fossils of morphologically advanced, foliose or fruticose lichens come from Tertiary (65–1.5 MaBP) deposits, i.e. are comparatively young; older ones certainly exist but have not yet been discovered. An easily recognizable impression of a Lobaria thallus (resembling L. pulmonaria) was found in early to middle Miocene deposits (24–12 MaBP) of a humid conifer forest from Trinity County, California (MacGinitie 1937; Peterson 2000), a site which might harbour many more lichen fossils (Peterson 2000). Amber, fossil- ized tree resins from the Old World (Baltic amber; 55–35 MaBP) and New World (Dominican amber; 20–15 MaBP) contain extremely well-preserved organisms. Most investigators focused on vertebrate, invertebrate or higher plant fossils, but also a few well-preserved lichens were detected (Ma¨gdefrau 1957). Recently two species of calicioid lichens were found in Baltic amber (Rikkinen 2003), and beautifully preserved thalli of two Parmelia species in Dominican amber (Poinar et al. 2000).
3.1.3 Cytological aspects There is no evidence for any fundamental difference between liche- nized and nonlichenized fungi. Cell wall structure and composition of lichen- forming ascomycetes occur within the range of variation observed in all Ascomycota (Honegger and Bartnicki-Garcia 1991). Lichen-forming ascomycetes 32 R. Honegger
and basidiomycetes produce and secrete the same type of hydrophobic cell wall surface compounds, the hydrophobins, as nonlichenized fungal taxa (see Chapter 5; Scherrer et al. 2000; Trembley et al. 2002a, b). For quite a while lichen-forming ascomycetes were supposed to differ from nonlichenized taxa by the possession of concentric bodies, semicrystalline cell organelles (Figs. 3.6, 3.7) of approximately 0.3 mm diameter, comprising a protein- aceous, electron-dense shell around a gas-filled center. Concentric bodies are not membrane-bound and usually occur in clusters near the cell periphery. They are neither a peculiarity of lichen-forming ascomycetes nor of the symbiotic way of life, since they were found in a range of plant pathogens and saprobes, some of which occur in climatically extreme habitats in microbial commu- nities of desert varnishes (Honegger 1993, 2001, 2006). Concentric bodies were found in the cytoplasm of all types of vegetative cells, in paraphyses, ascogen- ous hyphae, asci, and in mature ascospores. Longevity and a considerable desiccation tolerance seem to be the features shared by all ascomycetous cells that harbor concentric bodies; these were hypothesized to be remains of drought stress-induced cytoplasmic cavitation events (see Chapter 4), as regularly experienced by long-living fungal structures that are subjected to continuous wetting and drying cycles (Honegger 1995, 2001, 2006;Honegger et al. 1996). Multiperforate septa, a rather unusual feature among ascomycetes, have been observed even by early light microscopists and later with electron micro- scopy techniques (Wetmore 1973) in medullary hyphae of some Peltigeraceae and Parmeliaceae (Figs. 3.8, 3.9). Only some, but not all, medullary hyphae of a thallus reveal this structural peculiarity. It remains unknown whether these particular hyphae are functionally different from the rest of the medullary hyphae.
3.1.4 Symbiotic versus aposymbiotic phenotypes Aposymbiotic phenotypes in culture Lichen-forming fungi differ from nonlichenized taxa by their nutri- tional strategy and by their manifold adaptations to the cohabitation with a population of minute photobiont cells. In pure culture aposymbiotic lichen mycobionts form thallus-like colonies with no morphological resemblance to the symbiotic phenotype (Figs. 3.4, 3.5). The central, microaerobic part of such aposymbiotic thalli is usually composed of cartilaginous, conglutinate cell masses while filamentous growth, i.e. aerial hyphae, are formed at the periph- ery (Fig. 3.5; further examples in Ahmadjian, 1973; Honegger and Bartnicki- Garcia, 1991; Stocker-Wo¨ rgo¨ tter, 1995). Mycobionts 33
1 4 1 mm
1 mm
2 1 mm
35
uc
ph
m
lc 10 µm 10 µm
Figs. 3.1–3.5 Symbiotic and aposymbiotic phenotypes of the dorsiventrally organized macrolichen Xanthoria parietina (Teloschistales). Fig. 3.1. Laminal view, and Fig. 3.2. vertical cross section of the marginal lobes of the leaf-like thallus. Fig. 3.3. Detail of a vertical cross section: uc, conglutinate upper cortex; ph, photobiont layer harboring the globose cells of the green alga Trebouxia arboricola; m, gas-filled medullary layer built up by aerial hyphae; lc, conglutinate lower cortex. Fig. 3.4. Aposymbiotic phenotype on an agar medium. Fig. 3.5. Detail of the peripheral part of a cross section of the thallus-like fungal colony showing filamentous hyphal growth at the periphery and conglutinate zones in the microaerobic central part.
Symbiotic phenotypes in nature Most biologists consider lichens as being shaped like leaves (foliose; Fig. 3.1) or tiny, erect or pendulous shrubs (fruticose) and having an interesting internal stratification with the photobiont cell population incorporated in a thalline layer (e.g. Figs. 3.3 or 3.14). However, only one out of four lichen- forming fungi has such an impressive morphogenetic capacity. The majority of lichen mycobionts (Table 3.2) overgrow or ensheath photobiont cells on or 34 R. Honegger
6 7
0.5 µm 0.5 µm
8 9
1 µm 1 µm
Figs. 3.6–3.9 Figs. 3.6–3.7. Concentric bodies, semicrystalline cytoplasmatic orga- nelles of unknown origin and function in an ultrathin section and a freeze-fracture preparation of Peltigera canina (Peltigerales). Figs. 3.8–3.9. Multiperforate septa in medullary hyphae as observed in an ultrathin section of Parmelia tiliacea (Lecanorales) and a SEM preparation of Peltigera canina.
within the substrate and form microfilamentous (Fig. 3.13), microglobose (Fig. 3.10), or crustose thalli (Figs 3.11, 3.12), some of which are quite inconspic- uous. About 20% of the lichen-forming fungi form squamulose or placodioid thalli with an internal stratification, but usually these thalli remain in close contact with the substrate. In only about 25% of lichen species do the myco- bionts grow above the substrate and enter the third dimension by differentiat- ing either a foliose or fruticose thallus with internal stratification (Figs. 3.3, 3.14; Table 3.2). These morphologically complex symbiotic phenotypes are formed by a range of functionally and morphologically different fungal cells (see Chapters 4 and 5). The photobiont cell population is housed, maintained, and controlled within the fungal thallus. It is arranged similarly to the palisade parenchyma (i.e. the photosynthetically most active parts) in vascular plants: either in a plane, as in foliose lichens (Figs. 3.2, 3.3), or at the periphery of either erect (e.g. reindeer Mycobionts 35
1011 12
ph
ac
1 mm 1 mm 10 µm
13 14
cst
ph my c m
ph
5 µm 50 µm
Figs. 3.10–3.14 Structural and taxonomic diversity in lichen-forming fungi. Fig. 3.10. Omphalina ericetorum (Agaricales), a lichenized basidiomycete growing on detri- tus. Arrows point to microglobular lichenized structures (vivid green in fresh sam- ples) on the surface of a decaying leaf. Figs. 3.11–3.12. Graphis elegans (Graphidales) produces its crustose, grayish thallus within the smooth bark of Ilex europaeus. ac, lirelliform ascomata; ph, coccoid green algal photobiont cells (Trentepohlia sp.). Fig. 3.13. Microfilamentous thallus of the tropical Coenogonium subvirescens (Gyalectales). The filamentous green-algal photobiont (Trentepohlia sp.) (ph) is ensheathed by mycobiont hyphae (my). Fig. 3.14. Cross section of Usnea rubicunda (Lecanorales), a radially organized, fruticose lichen with internally stratified thallus. c, conglutinate cortical layer; ph, photobiont cell population (Trebouxia sp.); m, gas-filled medullary layer; cst, conglutinate central strand. 36 R. Honegger
lichens) or pendulous, radially organized structures (e.g. beard lichens; Fig. 3.14) analogous to the vegetative body of a variety of plants (e.g. Ephedra,rushes, many succulent plants in the Cactaceae, Stapeliaceae, etc.). In contrast to all other mutualistic symbioses of fungi and photoautotrophs, it is the fungal partner of these morphologically highly evolved lichens that secures adequate illumination and facilitates gas exchange of the photobiont (Honegger 1991a, 1992).
3.2 Specialized ‘‘lifestyles’’ of lichens and associated fungi
3.2.1 Parasitic lichens Parasitic lichens are a group of obligately lichenized fungi that start their development on or within the thallus of another lichen species. Depending on the degree of colonization, the host thallus is either locally or completely overgrown and destroyed. Accordingly, parasitic lichens may either be confined to host thalli throughout their lifetime or outgrow their host and become independent. In the former case the species are readily recognized as parasites, but in the latter case, extensive observations on all developmental stages may be required before the species is recognized as a parasite. Parasitic lichen-forming fungi acquire their photobiont either ‘‘by theft’’ from the host lichen or separately. In the latter case the compatible photobiont of the parasite is usually taxonomically not identical with the photoautotrophic partner of the host lichen (Poelt and Doppelbauer 1956; Hawksworth 1988b). A complicated mode of photobiont acquisition was observed in the crustose lichen Diploschistes muscorum. This parasitic lichen starts its development in the thallus squamules and podetia of Cladonia spp. (Fig. 3.18) that may be completely overgrown and destroyed. Juvenile D. muscorum associate with Trebouxia irregu- laris, the photobiont of the host lichen. However, large, independent thalli of D. muscorum were invariably found to have replaced Trebouxia irregularis by T. showmanii (Friedl 1987).
3.2.2 Bryophilous and foliicolous lichens A large number of crustose lichens favor decaying bryophytes as a substrate (e.g. the Buellietum olivaceobruneae in the Antarctic; Kappen 1985), but a small, taxonomically heterogeneous group (approx. 40 species) is parasitic on live bryophytes that are overgrown or even intracellularly infected (Figs. 3.15, 3.16; Poelt 1985). Compatible photobionts are often found under the cuticle between leaf and stem cells (Fig. 3.16) or, in some combinations, even within leaf cells (Do¨ bbeler and Poelt 1981). Mycobionts 37
15 17
ac
1 mm 1 mm
16 18 1 mm c c
lc
ph
st ac
10 µm
Figs. 3.15–3.18 Examples of multiple symbioses. Figs. 3.15–3.16. Dimerella lutea (Gyalectales), a bryophilous lichen developing its inconspicuous crustose thallus between the cuticle (c) and leaf cells (lc) of the foliose liverwort Frullania dilatata. Leaf and stem cells of the hepatic are invaded by the lichen mycobiont (arrows). ac, ascomata (pale orange); ph, coccoid cells of the green-algal photobiont, a Trentepohlia sp.; st, central strand. Fig. 3.17. Biatoropsis usnearum (Tremellales), a cecidogenous lichenicolous fungus induces gall formations (arrows) on the fruticose thallus of Usnea cornuta. Fig. 3.18. The parasitic lichen Diploschistes muscorum starts its develop- ment in the thallus squamules and cup-shaped podetia of Cladonia pyxidata and acquires its green-algal photobiont (Trebouxia irregularis) by theft from the host lichen (for details see Friedl, 1987). ac, ascomata of the parasitic lichen.
In tropical and subtropical areas, the diverse, taxonomically heterogeneous group of foliicolous lichens (c. 600 species; Farkas and Sipman 1993, 1997) has attracted considerable interest in recent years. Their crustose, microfilamen- tous or squamulose thalli develop, usually quite unspecifically, on perennial leaves of a very wide range of vascular plants from sea level to montane areas, 38 R. Honegger
with their greatest diversity being found in humid submontane rain forests (Gradstein and Lu¨ cking 1997;Lu¨ cking et al. 2003). Foliicolous lichens are bioin- dicators of microclimate (Lu¨ cking 1997). In one hectare of a submontane forest in Costa Rica c. 200 species of foliicolous lichens were detected (Gradstein and Lu¨ cking 1997). Because large numbers of economically important crops, such as many spice-producing shrubs and trees, Camellia (tea), various Citrus spp., Coffea, Hevea (rubber), Theobroma (cacao), etc., are colonized, these foliicolous lichens have attracted the interest of plant pathologists (Hawksworth 1988c). Most foliicolous lichen mycobionts are symbiotic with filamentous green algae of the genera Cephaleuros and Phycopeltis (both Trentepohliales, Ulvophyceae). Such algae occur abundantly in the aposymbiotic state on perennial leaves and are also considered as pests (Hawksworth 1988c). These potential photobionts grow either on the cuticle or below. Some of them even occur in the palisade par- enchyma of the leaf. In association with lichen mycobionts, the growth rate of these foliicolous algae is reduced. Most epicuticular lichens probably use the leaf merely as a substrate and grow equally well on artificial substrates such as plastic tape or slides (Sipman 1994; Sanders 2001a; Sanders and Lu¨ cking 2002), but subcuticular ones are quite likely to benefit from nutrients of vascular plant origin.
3.2.3 Lichenicolous fungi and fungal endophytes of lichen thalli A considerable number of fungi (approx. 1250 species in about 280 genera of ascomycetes and approx. 62 species in 10 genera of basidiomycetes; Hawksworth 1982, 1988b; Lawrey and Diederich 2003) gain their nutrition from lichens and a formal classification of relationships has been developed by Rambold and Triebel (1992). Beside these approx. 1300 lichenicolous species there are about 260 doubtfully and/or infrequently lichenicolous taxa. Some lichenicolous fungi are necrotrophic, i.e. have a devastating effect on either the lichen mycobiont (mycoparasites; Diederich 1996; de los Rı´os and Grube 2000), or on the photobiont (algal parasites; Grube and Hafellner 1990) or on both (e.g. the widespread basidiomycete Athelia arachnoidea). Others sporulate abundantly without causing major damage to either the fungal or the photoautotrophic partner of the host thallus. In the literature the former group is referred to as parasites, the latter as ‘‘parasymbionts’’ or ‘‘commensalists.’’ However, as very oligotrophic heterotrophs, these non-destructive inhabitants of lichens drain their nutrition from the thallus and thus are best regarded as mild parasites. Because experimental data are missing, it is often very difficult to interpret the biology of these multiple symbioses. With light and electron microscopic techniques it is often impossible to distinguish foreign hyphae within lichen thalli from the mycobiont proper. A few lichenicolous fungi are quite catholic Mycobionts 39 with regard to host preference (i.e. they have been found on different, unrelated taxa). However, the majority of lichenicolous fungi (an estimated 95%) shows a remarkable host specificity (Clauzade and Roux 1976; Hawksworth 1983; Lawrey and Diederich 2003). A group of approximately 40 species of lichenico- lous ascomycetes and basidiomycetes is particularly interesting because they induce more or less species-specific gall formations on their host thalli (Fig. 3.17). Some of these cecidogenous (gall-inducing) taxa stimulate, in an unknown manner, the growth of the mycobiont and photobiont (Triebel and Rambold 1988; Hawksworth and Honegger 1994); others parasitize the photo- biont, as observed in the heterobasidiomycete Biatoropsis usnearum (Grube and de los Rı´os 2001). The majority of lichenicolous fungi is likely to be exclusively lichenicolous, but a considerable number are also known as saprobes or as parasites of non- symbiotic aerophilic algae (Hawksworth 1982). Transitions between modes of nutrition have been recorded. An example is Chaenothecopsis consociata (Mycocaliciales), a lichenicolous fungus that invades thalli of Chaenotheca chryso- cephala (Caliciaceae, Lecanorales; photobiont: Trebouxia simplex), but, if available, it associates with Dictyochloropsis symbiontica to form its own crustose thallus (Tschermak-Woess 1980). The widespread heterobasidiomycete Athelia arachnoi- dea, necrotrophic on various lichen taxa, free-living algae and bryophytes especially in areas with severe air pollution (Yurchenko and Golubkov 2003), was shown to be the sexual state of Rhizoctonia carotae, a postharvest disease of carrots (Adams and Kropp 1996). Other Athelia spp. are lichenized (Oberwinkler 2001). Finally, there is a wide range of very inconspicuous, symptomless inhabitants of lichen thalli that were discovered by culturing experiments only (Petrini et al. 1990; Miadlikowska et al. 2004a, b). Among these fungal endophytes of lichen thalli are taxa that occur also as saprobes outside lichens. Others are known as endophytes or pathogens of higher plants. In addition, a range of partially characterized species has so far not been found outside lichen thalli (Petrini et al. 1990). The lichen thallus as an ecological niche for fungi not yet known to science is an interesting field for future research. 4 Thallus morphology and anatomy
b. b del and c. scheidegger
Symbiosis is now widely accepted as a source of evolutionary innovation (Margulis and Fester 1991) that has stimulated an enormous morphological radia- tion in ascomycetes. Vegetative structures have especially developed to a com- plexity that is not reached elsewhere in the fungal kingdom (Honegger 1991b). Lichen morphology and anatomy are now understood as being highly adapted to constraints imposed by the environment on the mutualistic symbiosis, where the mycobiont is the exhabitant and the cyanobacterial or green-algal photobiont is the inhabitant (Hawksworth 1988b). A very wide range of different thallus struc- tures have been described and a complete outline of lichen morphology is not the scope of this chapter. However, detailed reviews are given by Henssen and Jahns (1973) and Jahns (1988). Common mycological terms also used in lichenology are not always explained here. Readers are referred to recent mycological textbooks, to Hawksworth et al.(1983), or to a glossary of a recent lichen flora. Irrespective of lichen growth form, it must function as a photosynthetically active unit in a manner that allows positive net photosynthesis and subsequently sufficient growth rates. This implies that the photobiont has to be supplied with just the right amount of light, even in the deep shade of rain forests or under fully exposed conditions of deserts. Carbon dioxide (CO2) diffusion to the photobiont needs to occur readily, even under fully hydrated conditions. Water loss should be adapted to the specific environment: minimized in dry environments, and maximized in very wet environments. Thereby optimal CO2 gain may be realized.
4.1 Growth forms
The appearance of a lichen thallus is primarily determined by the mycobiont. Only a few cases are known where the photobiont determines the
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. Thallus morphology and anatomy 41 habit of the whole thallus, e.g. in the filamentous genera Coenogonium, Ephebe, Cystocoleus, and Racodium. However, knowledge of the influence of the photo- biont on the lichen morphogenesis is important, because only after the estab- lishment of the symbiosis is the characteristic thallus of a lichen developed. On the basis of their overall habit, lichens are traditionally divided into three main morphological groups: these are the crustose, foliose and fruticose types. There are a number of additional special types, such as the gelatinous lichens that all have cyanobacteria as photobionts (but not all lichens with cyanobacteria are gelatinous lichens!). However, all of them can be integrated within the threefold scheme of the main growth types.
4.1.1 Crustose lichens Crustose lichens are tightly attached to the substrate with their lower surface and may not be removed from it without destruction. Water loss is restricted primarily to the upper, exposed surface only. When growing on inclined rock surfaces, they profit from surface water flow. These features allow these organisms to tolerate extreme habitats such as bare, exposed rock surfaces. Although the crustose growth type seems to be clearly defined, varia- tion of the basically crustose type is abundant. The following subtypes can be distinguished: powdery, endolithic, endophloeodic, squamulose, peltate, pulvi- nate, lobate, effigurate, and suffruticose crusts. Their thallus organization may be either homoiomerous or heteromerous. In terms of complexity in thallus structure, the powdery crusts, as found in the lichen genus Lepraria, are the simplest and lack an organized thallus. Fungal hyphae envelop clusters of photobiont cells and do not have a distinct fungal or algal layer. They have a powdery appearance and are also referred to as leprose. Even more simple is the construction of thalli of the epiphyllic, epiphytic, and terricolous genus Vezdaea, in which the vegetative, photobiont-containing thal- lus consists of single, globose soredia-resembling granules, which are usually less than 1 mm in diameter. The granules occur either on the surface or under- neath the cuticle of bryophytes or other plant material. They are called gonio- cysts (Se´rusiaux 1985) and are corticate and often have distinct spines (Fig. 4.4). The construction of endolithic (growing inside the rock) and endophloeo- dic (growing underneath the cuticle of leaves or stems) lichens seems to be more organized. In most cases, an upper cortex is developed, e.g. in Lecidea aff. sarcogynoides (Wessels and Schoeman 1988). The upper cortex can consist of densely conglutinated hyphae forming a dense layer named ‘‘lithocortex’’, as for example in Acrocordia conoidea, Petractis clausa, Rinodina immersa, Verrucaria baldensis, and V. marmorea (Pinna et al. 1998). Other endolithic lichens like Verrucaria rubrocincta form a micrite layer with only a few hyphae involved 42 B. Bu¨ del and C. Scheidegger
(Bungartz et al. 2004). In Lecidea aff. sarcogynoides, the photobiont layer as a part of the medulla is located underneath the cortex within the rock. The medullary hyphae may extend up to 2 mm deep into the sandstone matrix. By penetrating the sandstone, L. aff. sarcogynoides weathers the sandstone at a rate of 9.6 mm per 100 years in the semihumid climate of South Africa (Wessels and Schoeman 1988). In contrast, Antarctic endolithic lichens apparently do not form a strongly stratified thallus (Friedmann 1982). In those lichens, thallus structure is mainly determined by the rock matrix itself (de los Rı´os et al. 2005). Active weathering by etching can be found in the marine species Arthopyrenia halodytes, living on calcareous substrates in the littoral. A. halodytes thalli can often be found in the shell of Balanus species (Fig. 4.1) and of molluscs. Crustose lichens (e.g. Leproplaca chrysodeta, Dirina massiliensis) contribute to the biodeter- ioration of a wide range of building materials and historical monuments (Nimis et al. 1992). Epilithic and epiphloeodic lichens comprise the vast majority of the crustose growth type and a number of special thallus types are developed. Most crustose lichens are stratified and show some, if not all, of the layers that are described under Section 4.2.2. The margin of the thallus may be clearly delimited or indistinct. In the genus Rhizocarpon or Placynthium, for example, a prothallus is developed. This is a photobiont-free, white or dark brown to black zone, visible between the areolae and at the growing margins of such crustose lichens. The corticolous lichen Cryptothecia rubrocincta in the rain forest of French Guiana collects water in its hydrophilic, nonlichenized prothallus and drains it through little channels passing underneath the thallus, thus avoiding periods of super- saturation. This was frequently observed after rainwater exuded continuously from experimentally induced injuries of the surface of the lichenized thallus part (Lakatos 2002; Lakatos et al. 2006). An areolated thallus consists of numerous areolae, which are polygonal parts of the thallus containing both symbiotic partners. In the dry state, the areolae are clearly distinguishable from each other, but under wet conditions they swell and the cracks between them close. The areolae might either develop from a primary thallus with a closed surface that splits secondarily from the surface to the bottom into several areolae, or from single thallus primordia developing on a prothallus. A thallus is called effigurate when the marginal lobes are prolonged and are radially arranged, as in many species of the genera Caloplaca, Dimelaena, Acarospora, and Pleopsidium. The squamulose type of crustose lichens is the most complex. The areolae are enlarged in their upper part and become partially free from the substrate. Often they develop overlapping scale-like squamules. This is the case, for exam- ple, in the genera Catapyrenium, Peltula, Psora, and Toninia. Flat scales of Thallus morphology and anatomy 43 squamulose thalli, with a more or less central attachment area on the lower surface, are called peltate (e.g. Peltula euploca, Anema nummularium; Fig. 4.2). The peltate type of squamulose crusts is often developed in lichens colonizing soils or rock surfaces in hot, arid regions of the world. Peltula radicata (Fig. 4.22), for example, colonizes soils in deserts of the world. It is completely immersed in the soil, exposing its flat upper surface only. Lichens with this type of growth were called ‘‘Fensterflechten’’ (i.e. window) by Vogel (1955), who first observed such lichens in the genera Eremastrella and Toninia. Extremely inflated squamules in the lichen genus Mobergia (Fig. 4.3) are called bullate. In other cases, coralloid tufted cushions, designated suffruticose, are formed. Within the genera Caloplaca and Lecanora,alobate thallus is developed by some species. This is the case when a thallus becomes radially striate with marginal, at least partially raised lobes.
4.1.2 Foliose lichens Foliose lichens are leaf-like, flat and only partially attached to the substrate. Foliose thalli are either homoiomerous (gelatinous lichens) or hetero- merous. Typically they have a dorsiventral organization with distinct upper and lower surfaces. Often the thallus is divided into lobes, which show various degrees of branching (Figs. 4.5–4.8). Foliose lichens develop a great range of thallus size and diversity. Laciniate lichens are the typical foliose lichens. They are lobate and vary considerably in size; they may either be gelatinous-homoiomerous (e.g. Collema, Leptogium, Physma; Fig. 4.8) or, as in most cases, heteromerous. The lobes can be radially arranged (e.g. in Parmelia species, Figs. 4.5, 4.6) or overlapping like tiles on a roof (e.g. Peltigera, Hypocoenomyce). In some genera, thallus lobes can become inflated, having a hollow medullary center (e.g. Menegazzia; Fig. 4.7). The lower surface is often covered by rhizinae, cilia or a tomentum, which, to a limited degree, may also serve as attachment structures. Conspicuous lichens, such as the genera Sticta and Pseudocyphellaria in the understory of tropical and tempe- rate rain forests, Lobaria and Nephroma in alpine and oceanic forests, or Peltigera in arctic tundras, belong to this growth type. Umbilicate lichens have circular thalli, consisting either of one single, unbranched lobe or multilobate thalli with limited branching patterns. All are attached to the substrate by a central umbilicus from the lower surface. This often can be recognized by a navel-like depression on the upper side. The umbilicus usually consists of tightly packed, parallely arranged and congluti- nated hyphae without photobiont cells. Umbilici have apparently evolved sev- eral times in such unrelated groups as the Dermatocarpaceae, Parmeliaceae, Physciaceae, and Umbilicariaceae. 44 B. Bu¨ del and C. Scheidegger
12
1mm 1mm
3 4
10 µm
Figs. 4.1–4.4 Growth forms, crustose lichens. Fig. 4.1. Arthopyrenia halodytes (Spain, Mediterranean Sea, on Balanus sp.), an endolithic lichen within calcareous substrates with perithecia (arrow). Fig. 4.2. Anema nummularium (Spain, on conglomerate), a gelatinous lichen with unicellular cyanobionts and lecanorine apothecia (white arrow); the thallus is squamulose-umbilicate and effigurate at the margins (arrowhead). Fig. 4.3. Mobergia calculiformis (Baja California, on rock), bullate thallus. Fig. 4.4. Vezdaea rheocarpa (Switzerland, over epiphytic mosses), goniocysts (SEM micrograph). Thallus morphology and anatomy 45
5 6
2mm
7 8 2mm
5 mm
Figs. 4.5–4.8 Growth forms, foliose lichens. Fig. 4.5. Parmelia pastillifera (Switzerland, on Acer pseudoplatanus), thallus with black, knob-like isidia. Fig. 4.6. Parmelia sulcata (Germany, on bark), flat thallus lobes with soralia (arrow). Fig. 4.7. Menegazzia terebrata (New Zealand, temperate rain forest, on Nothofagus men- ziesii), thallus with inflated, hollow lobes with pores (arrow) and slightly stalked, lecanorine apothecia (arrowhead). Fig. 4.8. Physma sp. (Australia, tropical rain forest, on bark), gelatinous foliose lichen. 46 B. Bu¨ del and C. Scheidegger
An ecologically interesting group of foliose lichens are the vagrant lichens, such as Xanthomaculina convoluta and Chondropsis semivirdis in deserts and semideserts. These lichens show hygroscopic movement (Bu¨ del and Wessels 1986;Lumbsch and Kothe 1988). In the dry state the thalli are rolled up, thus exposing their lower cortices. When they take up liquid water, the thalli unroll and expose the upper surface to the sunlight. In both lichens photosynthesis is considerably increased after exposing the upper surface to the light (Lange et al. 1990b). When dry and inrolled, the lichens can easily be blown by the wind and as soon as dewfall occurs, they unroll and expose the upper surface again.
4.1.3 Fruticose lichens The thallus lobes of fruticose lichens are hair-like, strap-shaped or shrubby and the lobes may be flat or cylindrical. They always stand out from the surface of the substrate. Some groups have dorsiventrally arranged thalli (e.g. Sphaerophorus melanocarpus, Evernia prunastri), but the majority possess radial symmetric thalli (e.g. Sphaerophorus globosus, Usnea species, Ramalina species; Figs. 4.11, 4.12). The branching pattern of lobes varies considerably among different systematic groups and also within a single genus. Size varies tremen- dously, from some Usnea species that grow several meters long to minute species only 1 or 2 mm high. Fruticose lichens are found in a wide range of climates, from the desert to the wet rain forest and on various types of substrates. Genera like Baeomyces or Cladonia (Fig. 4.10) develop a twofold thallus, which is differentiated into a fruticose thallus verticalis and a crustose-squamulose to foliose thallus horizontalis. In the genus Cladonia, the thallus verticalis origi- nates from primordia of the fruit body and results in apothecia-bearing stalks, which are termed podetia (Fig. 4.10). In cases where the thallus verticalis devel- ops from primordia of the thallus horizontalis, e.g. in the genus Stereocaulon,itis named pseudopodetium. The genus Cladia has developed reticulate podetia, thus exposing the medulla and parts of the algal layer (Fig. 4.9). As a consequence, uptake and loss of water is rapid. This feature probably reflects the wet soils and moss cushions within which many of the reticulate Cladia species grow. Another peculiar type of anatomy is that of the beard-like-growing genus Usnea (Fig. 4.12), which has a strong central strand of periclinally arranged, congluti- nated hyphae that provide mechanical strength along the longitudinal axis. Highly branched fruticose lichens have a high surface to volume ratio, which results in a more rapid drying and wetting pattern compared with lichens having lower surface to volume ratios. Fruticose growth forms can be found preferentially either in very wet, humid climates, e.g. Usnea xanthophana and U. rubicunda in the temperate rain forest, or in arid climates with regular dew and fog events, e.g. Teloschistes capensis in the Namib Desert. While Usnea Thallus morphology and anatomy 47
9 10
500 µm
11 12
1 mm
Figs. 4.9–4.12 Growth forms, fruticose lichens. Fig. 4.9. Cladia retipora (New Zealand, temperate rain forest, on soil), fenestrate lobes (SEM micrograph). Fig. 4.10. Cladonia coccifera (Russia, Lake Baikal, on soil), vertical thallus (¼ podetia) with red hymenium (arrow). Fig. 4.11. Ramalina pollinaria (Switzerland, on Abies alba), typical fruticose growth form. Fig. 4.12. Usnea filipendula (Germany, Fagus forest, on Fagus sylvatica), radial symmetric thallus with numerous fibrils. 48 B. Bu¨ del and C. Scheidegger
xanthophana and U. rubicunda hardly showed any depression in CO2-uptake due to
slower CO2-diffusion under water-supersaturated conditions (Lange et al. 1993b), Teloschistes capensis is able to collect sufficient fog and early morning dewfall to sustain positive net photosynthesis for considerable time periods (Lange et al. 1990a). In the former habitat, the hanging ‘‘beard’’ of the Usnea thalli is advanta- geous in that rapid water loss occurs. Together with the numerous hydrophobic airspaces in the medulla, supersaturation with water is avoided (Lange et al. 1993b). In the second case, the large tufts of Teloschistes capensis act like a comb for dew and fog condensation. In a similar manner, increased water uptake occurs in many other fruticose genera, such as Ramalina menziesii (Larson et al. 1985; Boucher and Nash 1990a) and the genera Bryoria and Usnea.
4.2 Vegetative structures
4.2.1 Homoiomerous thallus Mycobionts and photobionts are evenly distributed in homoiomerous thalli, as is often found in thin crustose lichens, in gelatinous crustose and foliose lichens, for example of the genera Caloplaca, Pyrenopsis or Collema (Fig. 4.13). The homoiomerous gelatinous lichens absorb much more water in relation to their dry weight than nongelatinous, heteromerous lichens do. As a conse-
quence, CO2 gas diffusion to the photobiont is strongly limited or may even be blocked in supersaturated thalli. Carbon dioxide can become the limiting factor for photosynthesis under these circumstances (Lange and Tenhunen 1981). However, Lange et al.(1993b) showed for Collema laeve and other cyanobacterial
lichens that CO2 uptake is only slightly depressed under supersaturated condi- tions, a fact that may be due to the functioning of the recently described
photosynthetic CO2 concentration mechanism. A steeper concentration gradi-
ent from air outside to the CO2 fixation site within the cyanobacterium leads to
an increasing diffusion rate of CO2 (Badger et al. 1993).
4.2.2 Stratified thallus The majority of lichens including many crustose species develop intern- ally stratified thalli. The main subdivisions are into upper cortex, photobiont layer, medulla, and lower cortex. These layers may include various tissue types, and their terminology follows the general mycological literature. In the case where the hyphae are conglutinated to an extent that single hyphae are not usually distinguishable, the main tissue types are pseudoparenchymatous and/or prosoplectenchymatous. The former case involves more or less isodia- metrical cells (Fig. 4.18) resembling true parenchyma of vascular plants. The latter Thallus morphology and anatomy 49
13 14
10 µm
10 µm
15 16
10 µm
10 µm
Figs. 4.13–4.16 Thallus anatomy, heteromerous versus homoiomerous thalli, water storage. Fig. 4.13. Collema nigrescens (Switzerland, on Acer pseudoplatanus), a wet, homoiomerous thallus of the gelatinous type with the filamentous Nostoc cyanobiont (upper arrow). Water is located in the jelly of fungal and cyanobacterial walls and 50 B. Bu¨ del and C. Scheidegger
tissue type is composed of elongated hyphal cells that are arranged anticlinally. For the description of dense plectenchyma in ascoma, a terminology introduced by Korf (1973) for nonlichenzed discomycetes is useful (Scheidegger 1993).
Cortex, epicortex, and epinecral layer Only a few growth forms are known where the photobiont layer imme- diately reaches the surface of the lichen thallus. Mostly the algal layer is covered by a thin to thick (up to several hundred micrometers) cortical layer (Fig. 4.14). In many dark lichens pigmentation is confined to fungal cell walls of cortical hyphae (Esslinger 1977;Timdal1991)ortheepinecral layer. In gelatinous lichens the color is primarily confined to the outer wall layers of the mycobiont (Bu¨del1990). In many foliose or fruticose lichens a cortex is formed by pseudoparenchyma- tous or a prosoplectenchymatous fungal tissue. Usually living or dead photobiont cells are completely excluded from the cortex, but in the so-called phenocortex (Poelt 1989) collapsed photobiont cells are included (e.g. Lecanora muralis). In Parmeliaceae some species have a 0.6 to 1 mm thick epicortex, which is a noncellular layer secreted by the cortical hyphae. This epicortex can be pored, as in Parmelina, or nonpored, as in Cetraria. In a broad range of foliose to crustose lichens an epinecral layer of variable thickness is often developed. It consists of dead, collapsed and often gelatinized hyphae and photobiont cells. Thalli often have a whitish, flour-like surface covering, the so-called pruina that consists primarily of superficial deposits, of which calcium oxalate is the most common. The amount of calcium oxalate is probably dependent on ecological parameters, such as calcium content of the substrate and the aridity of the microhabitat. Functions of the upper cortex and/or its pruina include mechanical protec- tion, modification of energy budgets (Kershaw 1985), antiherbivore defense (Reutimann and Scheidegger 1987), and protection of the photobiont against excessive light (Ertl 1951;Bu¨ del 1987; Jahns 1988; Kappen 1988). Light- and shade-adapted thalli of several species differ considerably in the anatomical organization of their upper cortical strata (e.g. Peltigera rufescens; Figs. 4.17,
Caption for Figs. 4.13–4.16 (cont.) sheaths (lower arrow). There are no airspaces in the thallus (LTSEM micrograph). Fig. 4.14. Anaptychia ciliaris (Switzerland, on Acer pseudoplatanus), a wet, heteromerous thallus with unicellular Trebouxia phycobionts in the algal layer (lower arrow). Water is located in the cell walls of the mycobiont mainly (upper arrow). Numerous air- spaces are located in the medulla (LTSEM micrograph). Fig. 4.15. Nephroma resupinatum (Switzerland, on Acer pseudoplatanus), a wet, heteromerous thallus with filamentous Nostoc cyanobionts (LTSEM micrograph). Fig. 4.16. Cetraria islandica (Sweden, terricolous), a dry, heteromerous thallus with Trebouxia photobiont. Note the collapsed phycobiont cells (LTSEM micrograph). Thallus morphology and anatomy 51
4.18). In a nearby fully sun exposed habitat, the thickness of the cortex is reduced but a thick, epinecral layer with numerous air spaces is formed (Fig. 4.17), giving the thallus a grayish-white surface due to a high percentage of light reflection (Dietz et al. 2000). This cortical organization results in decreased transmission of incident light by 40% in the sun-adapted thallus measured at the upper boundary of the algal layer. Additionally, epinecral layers of Peltula species contain airspaces that may also function as CO2 diffusion paths under supersaturated conditions (Bu¨ del and Lange 1994).
Photobiont layer and medulla In most foliose or fruticose thalli, the medullary layer occupies the major part of the internal thalline volume. Usually it consists of long-celled, loosely interwoven hyphae forming a cottony layer with a very high internal airspace. The upper part of the medulla forms the photobiont layer. In many lichens, the hyphae of the photobiont layer are anticlinally arranged and may sometimes form short or globose cells. Supporting tissue is often formed within the medullary layer of fruticose lichens and to a secondary degree in other lichens. It consists of thick-walled, conglutinated hyphae. This special tissue may be formed as irregularly arranged hyphal strands (e.g. in maritime, placodioid crustose lichens), as a central cylinder (Cladina) or as a central, thread-like elastic strand (Usnea). The hyphal cell walls of the algal and medullary layer are often encrusted with crystalline secondary products. These crystals and/or tessellate outer cell wall layers (Honegger 1991c) make the medullary hyphae hydrophobic. Therefore, during wet periods the medullary and algal layer remain air filled and capillary water is probably not present in internal parts of the thalli (Figs. 4.42, 4.43; Brown et al. 1987; Scheidegger 1994a). Water transport to the photobionts seems to be restricted to mycobiont cell walls. Under water-saturated conditions, photobiont and mycobiont cells are fully turgid (Figs. 4.14, 4.15). But in the air- dry state photobiont cells are collapsed following water loss (Fig. 4.16). Green- algal symbionts have the ability to become turgid (Brown et al. 1987;Bu¨ del and Lange 1991; Scheidegger 1994a) and achieve positive net photosynthesis at high relative air humidity ( >85 %) and low temperatures. However, cyanobacterial photobionts only become turgid in the presence of liquid water (Bu¨ del and Lange 1991; Scheidegger 1994a). Although some thin-walled medullary hyphae collapse during water loss, cortical and thick-walled medullary fungal hyphae usually show a different reaction to water loss. These cells cavitate and keep their shape more or less unaltered during the desiccation process (Figs. 4.37–4.39). Cavitation is an explosion-like formation of bubble-like struc- tures in the symplasts of fungal hyphae and spores (Scheidegger et al. 1995a). 52 B. Bu¨ del and C. Scheidegger
17 18
20 µm 20 µm
19 20
20 µm 20 µm
21 22
50 µm 1 mm
Figs. 4.17–4.22 Thallus anatomy, functional anatomy, vegetative structures. Figs. 4.17–4.18. Peltigera rufescens (Germany, xerothermic steppe, on soil, from Dietz et al. 2000). Fig. 4.17. Specimen from a sunny, fully exposed habitat, upper cortex with an additional epinecral layer (arrow). Fig. 4.18. Specimen from a permanently shaded habitat; the pseudoparenchymatic cortex is thicker compared with the fully exposed specimen and lacks an epinecral layer. Fig. 4.19. Sticta latifrons (New Zealand, tempe- rate rain forest), cross section with cyphellae with the typical corticated anatomy at Thallus morphology and anatomy 53
Cavitation bubbles take a major part of the volume of the symplast and compensate the volume of the water that has evaporated during desiccation. During rehydration, cavitation bubbles are refilled within seconds. Because cavitation does not change the shape of the cells, desiccation and rehydration processes that change water content of lichen thalli between <20% and >150% of the lichen dry weight do not induce shearing forces between fungal hyphae, for example in a multilayered cortex (Scheidegger et al. 1995a). Even in freezing and melting cycles of hydrated lichens the same morphological processes can be observed. In Umbilicaria aprina cooling of hydrated lichen thalli below the freezing point leads to extracellular freezing and an accumulation of ice crystals in the intercellular space of the lichen medulla. The effects of this freezing process are identical to dehydration and lead to collapsed photobionts and cavitated fungal hyphae (Schroeter and Scheidegger 1995). Free, capillary water has only been found thus far within the hollow podetia of Cladonia and Cladina species, in which an internal, hydrophilic central cylin- der is developed.
Lower cortex In some foliose lichens such as Peltigera or Heterodermia the medulla directly forms the outer, lower layer of the thallus. However, typical foliose lichens of the Parmeliaceae and many other groups have a well-developed lower cortex. As is the case with the upper cortex, it is either formed by pseudopar- enchymatous or a prosoplectenchymatous tissue. But unlike the upper cortex, the lower cortex is often strongly pigmented. Its ability to absorb water directly is well documented. Only low water conductance has been found thus far. However, it may play a major role in retaining extrathalline, capillary water (Jahns 1984).
Attachment organs and appendages An impressive variety of attachment organs may be developed from the lower cortex and also rarely from the thallus margin or the upper cortex. In
Caption for Figs. 4.17–4.22 (cont.) the lower surface. Fig. 4.20. Pseudocyphellaria filix (New Zealand, temperate rain forest), cross section with pseudocyphellae showing the typical protrusion of medullary hyphae into the opening (arrow). Fig. 4.21. Pseudocyphellaria dissimilis (New Zealand, temperate rain forest), cross section through the heteromerous thallus with Nostoc photobiont and with a tomentum forming hyphae originating from lower cortical cells. Fig. 4.22. Peltula radicata (Saharan Desert, soil), squamules connected with rhizines (arrow) to each other forming one thallus. 54 B. Bu¨ del and C. Scheidegger
foliose lichens attachment is mainly by simple to richly branched rhizines, mostly consisting of strongly conglutinated prosoplectenchymatous hyphae. Umbilicate lichens as well as Usnea and similarly structured fruticose lichens are attached to the substrate with a holdfast, from which hyphae may slightly penetrate into the substrate. Deeply penetrating rhizine strands are found in some squamulose, crustose or fruticose lichens growing in rock fissures, over loose sand and sod. In crustose lichens a prosoplectenchymatous prothallus is often formed around and below the lichenized thallus. It establishes contact with the sub- strate. From there bundles of hyphae penetrate among soil particles. Members of various growth forms produce a loose web of deeply penetrating hyphae, growing outwards from the noncorticate lower surface of the thalli. Various attachment organs with high amounts of extracellular gelatinous material establish tight contacts to the substrate. In this manner attachment to loose substrates, such as sand, is possible. Cilia are fibrillar outgrowths from the margins or from the upper surface of the thallus. A velvety tomentum consisting of densely arranged short, hair-like hyphae may be formed on the upper or lower cortex. Tomentose surfaces are mainly reported from broad-lobed genera such as Pseudocyphellaria, Lobaria, and Sticta but are also found in a few Leptogium and others.
Cyphellae and pseudocyphellae Upper or lower cortical layers often bear regularly arranged pores or cracks. Pseudocyphellae, as found on the upper cortex of Parmelia sulcata (Fig. 4.24) or on the lower side of Pseudocyphellaria (Figs. 4.20, 4.21), are pores through the cortex with loosely packed medullary hyphae occurring to the interior. Cyphellae are bigger and anatomically more complex than pseudo- cyphellae. In the interior portions of the cyphellae, hyphae form conglutinated, globular terminal cells, and this is the main difference from pseudocyphellae. They are only known from the genus Sticta (Fig. 4.19). Pseudocyphellae and young cyphellae are thought to considerably lower gas diffusion resistance of the cortex. Because pseudocyphellae and cyphellae are hydrophobic structures, they may act as pathways for gas diffusion into thalli. However, under supersaturated conditions they can no longer function in this way (Lange et al. 1993b).
Cephalodia (Photosymbiodemes) Representatives of the foliose Peltigerales with green algae as the pri- mary photobiont, and members of such genera as Stereocaulon, Amygdalaria, Chaenotheca, Micarea,andPlacopsis (Fig. 4.25) usually possess an additional 23 24
50 µm 50 µm
25 26
50 µm
Figs. 4.23–4.26. Vegetative structures and anamorph structures. Fig. 4.23. Parmelia crinita (Switzerland, saxicolous), thallus with finger-like isidia (LTSEM micrograph, courtesy of S. Geissbu¨ hler). Fig. 4.24. Parmelia sulcata (Switzerland, on Acer pseudoplatanus), soralium with numerous soredia (LTSEM micrograph). Fig. 4.25. Placopsis gelida (Sweden, saxicolous), thallus with a centrally arranged, gall-like external cephalodium, containing the N-fixing cyanobiont. Fig. 4.26. Heppia lutosa (Arizona, Sonoran Desert, terricolous), cross section exposing the immersed, flask-shaped pycnidium with pycnospores. 56 B. Bu¨ del and C. Scheidegger
cyanobacterial photobiont. In Solorina this secondary photobiont may form a second photobiont layer underneath the green-algal layer, but usually it is restricted to minute to several millimeters wide cephalodia. Cephalodial morphology is often characteristic on a species level and ranges from internal verrucae to external warty, globose, squamulose or shrubby structures on the upper or lower thallus surfaces. Cephalodial morphology usually differs completely from the green-algal thallus, and this emphasizes the potential morphogenetical influence of the photobiont on the growth form of the mycobiont–photobiont association (Chapter 5). Because many cyanobacterial photobionts are nitrogen-fixing (Chapter 11), these lichens may considerably benefit from cephalodia, especially in extremely oligotropic habitats.
4.3 Reproductive structures
As occurs in most fungi, the vast majority of lichenized ascomycetes have a sexual and an asexual life cycle. Within lichens, usually only the myco- biont expresses the full sexual and, to a certain degree, also asexual reproduc- tion. The reproduction mode of the photobiont is, however, reduced in the lichenized state. The principal problem with lichenization is the necessity of fungal spores meeting the proper photosynthetic partner for the re-establishment of the symbiosis. In addition to the typical sexual (teleomorph) and asexual (anamorph) fruiting structures of the individual symbionts, lichenized asco- mycetes have evolved a number of vegetative propagules, by which both part- ners are distributed.
4.3.1 Generative reproduction: ascoma (pl. ascomata) In contrast to the vegetative thallus, ascomata are composed of haploid hyphae and dicaryotic ascogenous hyphae. Two main ascocarp development lines are distinguished in the Euascomycetidae. In the ascolocular type, asci arise in cavities in a preformed stroma. True paraphyses and an excipulum proprium (proper or true margin) are lacking. Many lichenized fungi with the ascolocular type have fruit bodies that resemble perithecia, but because of their ascolocular origin, they are called pseudothecia (e.g. Pleosporales, Hysteriales). The second and most common ascocarp development in lichenized ascomy- cetes is the ascohymenial type, which is initiated with the development of an ascogonium, followed by the establishment of dicaryosis. It is still unknown Thallus morphology and anatomy 57 how dicaryosis occurs in lichenized ascomycetes. The ascocarp is composed of ascogenous hyphae and haploid hyphae from the base of the ascogonia-bearing hyphae and the asci are developed from ascogenous hyphae. Together with the sterile paraphyses (hamathecium), they form the hymenium. Underneath the hymenium a generative layer (i.e. subhymenium) is present, which gives rise to the hymenium. Sometimes a hypothecium, underlying the subhyme- nium, can be developed. The hymenium itself may be overlaid by a distinct epithecium.Thedevelopmentofafruitbodyisagymnocarp, when the hymenium is exposed from the earliest stage on. An initially enclosed fruiting body that opens before forming a fully mature hymenium is typical of hemi- angiocarp development. If the hymenium remains closed until the spores are mature, then it is called angiocarp development. For more detailed descriptions the reader is referred to basic mycological and lichenological works. There is considerable variation among ascomata and according to their morphology and anatomy several types of fruit bodies exist. In apothecia the hymenium is exposed at maturity and the hamathecium is either lacking or consists of paraphyses, paraphysoids or pseudoparaphyses; whereas perithecia remain closed, not exposing the hymenium.
Perithecia Perithecia open with a small tube-like ostiolum and have peri- physes and sometimes paraphyses (hamathecium). They are globose to flask-- shaped and are more or less immersed (e.g. Arthopyrenia, Catapyrenium, Dermatocarpon; Fig. 4.1). The exciple is carbonized in some genera, as in Verrucaria, and the ostiolum may be surrounded by a shield-like, carbonized layer.
Apothecia Apothecia are cup- or disk-shaped (Figs. 4.2, 4.7, 4.8) and two main morphological and developmental types are distinguished. Apothecia with a margin originating from the thallus (margo thallinus) are lecanorine (Fig. 4.31). In other cases where the margin develops from the tissue of the fruit body (margo proprius), it is either called lecideine (Fig. 4.32) when the margin is carbonized or biatorine when noncarbonized. In some genera, both margins are present, and this type is called zeorin. Within the margo proprius, two layers can be distinguished: the inner part is formed by the parathecium, giving rise to the outer layer, the amphithecium. 58 B. Bu¨ del and C. Scheidegger
Thallinocarpia Thallinocarpia are derived from a strongly, if not totally, reduced gen- erative tissue. In this special type, the ascogonia develop from thallus hyphae underneath the lichen surface, between mycobiont and photobiont cells. Subsequently, generative tissue is developed, from which true paraphyses are formed (Henssen et al. 1981). Parts of the thallus, including photobiont cells, are dispersed in between the hymenial parts so that the fruiting structure is sub- divided into thalline and generative compartments (Fig. 4.33). Thallinocarpia are only known from the family Lichinaceae, e.g. the genus Lichinella and Gonohymenia.
Pycnoascocarpia Pycnoascocarpia are a special ontogenetical type of apothecia, originat- ing by transformation of pycnidia into apothecia. They occur in the genera Ephebe, Paulia,orThyrea (Henssen et al. 1981).
Hysterothecia Hysterothecia are elongated, small fruit bodies with a split-like hyme- nium (Fig. 4.34). They may be either derived from perithecia or, as in all liche- nized ascomycetes, from apothecia. Hysterothecia are found in the genera Graphis and Opegrapha.
Asci The ascus structure and function plays an important role in ascomycete systematics. On the basis of electron microscopical and traditional light micro- scopical investigations, it is evident that characteristic types of asci occur within the lichenized ascomycetes. A basic difference between ascus types seems to be prototunicate asci and unitunicate/bitunicate ones. These latter two terms are derived from the original light microscopy studies, from which walls were thought to be either one- or two-layered. However, electron-microscopical stu- dies have subsequently demonstrated that this terminology is misleading. To reduce confusion we distinguish here between anatomically and functionally single- or two-layered asci. Prototunicate asci have anatomically and functionally single-layered, thin walls without any special mechanisms for spore dehiscence. Spores are either released by apical splitting or disintegration of the wall (Figs. 4.35, 4.36H, I). From an anatomical perspective unitunicate asci have a two-layered wall, but it behaves functionally as a single layer. Spore release is usually supported by an apical apparatus (part of the inner wall), which shows a high degree of Thallus morphology and anatomy 59
27 28
100 µm 5 µm
29
1 µm
30
1 µm
Figs. 4.27–4.30 Conidia. Fig. 4.27. Micarea adnata (Switzerland, on Abies alba), sporodochium with conidiospores (LTSEM micrograph). Fig. 4.28. Amandinea coniops (Norway, saxicolous) conidiogenous cells with filiform conidia (SEM micrograph). Figs. 4.29–4.30. Parmelia tiliacea, part of conidiogenous cell with enteroblastic formation of the primary conidium. A collarette is seen at the neck of the phialide. TEM (Fig. 4.29) and SEM (Fig. 4.30) micrographs (from Honegger 1984a, with permission from the author and publisher). 60 B. Bu¨ del and C. Scheidegger
31 32
100 µm 50 µm
33 100 µm 34
35
1 µm 10 µm
Figs. 4.31–4.35 Types of ascomata. Fig. 4.31. Pyxine physciaeformis (Brasil, epiphytic), lecanorine apothecium with algal cells in the excipulum (arrow; SEM micrograph). Fig. 4.32. Buellia leptocline (Sweden, saxicolous), section of the excipulum of the lecideine apothecium (from Scheidegger 1993). Fig. 4.33. Lichinella intermedia (Baja California, volcanic rock), typically cracked thallinocarp. The hymenium is divided Thallus morphology and anatomy 61 variability (Fig. 4.36C–G) and seems to be of great value in systematics. Prior to spore release, the outer wall opens and the apical apparatus elongates towards the surface of the hymenium. This type is termed nonfissitunicate, because the wall layers do not split (Honegger 1982b). Bitunicate asci have two wall layers that also function like two layers. The outer wall (exoascus) is not expandible and opens apically. The inner wall (endoascus), which is at least partly liberated from the exoascus (Fig. 4.36A, B), is highly expandible and elongates substantially towards the hymenial surface, prior to spore release. This type is termed fissitunicate because the inner, expandible wall is liberated from the outer nonexpandible wall (Honegger 1982b). Variation of the two basic types (fissitunicate and nonfissi- tunicate) have undergone evolutionary radiation. The systematics chapter (Chapter 17) in this book demonstrates the use of ascus structure as a phylo- genetic marker.
Basidioma (pl. basidiomata) To date only a few species of the basidiolichenes are known and none forms a specific lichen thallus. All basidiolichens belong to the Holobasidiomycetidae, which is characterized by nondivided basidia. The fruit- ing structures are typical of the Aphyllophorales or Agaricales. They are either crust-like (resupinate, e.g. Dictyonema), cylindrical or club-shaped (clavarioid, e.g. Multiclavula) or have the shape of a lamellated mushroom fruit body (agaricoid, Omphalina). For further information see Oberwinkler (1984).
4.3.2 Vegetative reproduction Aposymbiotic (apo ¼ non) propagules Conidiomata occur in many lichenized ascomycetes. Pycnidia are the main type of anamorph structures and are pear-shaped or globose receptacles (Fig. 4.26), within which conidia are formed on a special hyphal type, called conidiophores (Figs. 4.28, 4.29, 4.30). Pycnidia typically have a plectenchyma- tous wall. Several types of pycnidia are distinguished by the arrangement and morphology of the conidiophores (Vobis and Hawksworth 1981; Hawksworth 1988d). Conidiogenesis seems to be phialidic in all cases investigated so far
Caption for Figs. 4.31–4.35 (cont.) into numerous small partial hymenia that are interrupted by photobiont-containing, vegetative thallus parts (from Henssen et al. 1986). Fig. 4.34. Vezdaea aestivalis (Switzerland, over saxicolous bryophytes), ascus with covering paraphysoids. (LTSEM micrograph, from Scheidegger 1994b). Fig. 4.35. Graphis scripta (Germany, on Fagus sylvatica), hysterothecia with a letter-like shape. 62 B. Bu¨ del and C. Scheidegger
Fig. 4.36 Ascus types before (left) and after (right) dehiscence: A, Peltigera-type; B, Rhizocarpon-type; C, Euopsis-type; D, Lecanora-type; E, Pertusaria-type; F, Teloschistes-type; G, Anzina-type; H, Heppia-type; I, Chaenotheca-type. The horizontal line indicates the surface of the hymenium. Chaenotheca has a mazaedium (i.e. the ascus is covered by a thick layer of spores). ac, apical cushion; ar, eversible amyloid ring; aw, ascus wall; d, apical dome; e, electron dense particles; il, inner layer; ol, outer layer; p, preformed porus. They are redrawn from Honegger’s (1982b) TEM micrographs (A,B,D,E,F); Henssen et al. 1987 (C); Scheidegger 1985 (G); Bu¨ del 1987 (H); and Honegger 1985 (I) with permission from the authors.
(Figs. 4.28, 4.29, 4.30). The shape of the conidia, their mode of formation and the construction of the receptacles often have taxonomic value and may be used in characterizing species or even genera. Campylidia are erect, helmet-shaped conidiomata known from several foliicolous lichens. In the Badimia pollilensis aggregate, campylidia develop from primordia identical to those of the apothecia (Se´rusiaux 1986). Only one lichen (Micarea adnata) is known where the anamorph is a spor- odochium. Sporodochia are characterized by palisade-like arranged conidio- phores, developing on or reaching the surface of the lichen where they release conidia directly (Fig. 4.27). Thallus morphology and anatomy 63
Hyphophores areaspecialanamorphstructureknownfromthemyco- biont of many foliicolous lichens. They have a multihyphal stalk and often an apical thickening, from which conidiophores grow downwards (Ve˘ zda 1979). Thallospores are special asexual spores having no conidiophore. They are known from Umbilicaria (Hasenhu¨ ttl and Poelt 1978; Hestmark 1990), as well as from various crustose species (Poelt and Obermayer 1990a). Propagation of the aposymbiotic photobionts is seemingly rather rare (Tschermak-Woess 1988). Flagellate stages of unicellular green algae have been reported from Flavoparmelia caperata (Slocum et al. 1980)andfrom Anzina carneonivea (Scheidegger 1985). However, motile stages of the photo- bionts are usually absent or very rarely found in lichens, although they are regularly found in the free-living, cultured state of algae (Tschermak-Woess 1988). Hormogonia are short chains of cyanobacterial cells that act as diaspores. The cyanobacterial photobiont of Placynthium nigrum forms hormogonia after periods of high rainfall, and these may escape the lichen thallus (Geitler 1934).
Symbiotic propagules Vegetative diaspores provide a means of propagating the whole lichen- ized ascomycetes and basidiomycetes without relichenization. A wide range of ontogenetically and/or structurally different vegetative propagules have been described (Poelt 1993). However, only a few are routinely considered, and of these, isidia and soredia are the most important (Table 4.1). In most lichens, fragments are not normally able to establish and regen- erate a thallus. However, many fruticose lichens are highly adapted to propa- gate by thallus fragments. Beard-like, epiphytic thalli of the genera Bryoria and some Ramalina are torn and dispersed by strong winds. The fragments with lengths often exceeding 10 cm get entangled with the foliage and branches of their new substrate. Some Cladonia species are very brittle in the dry state and are often fragmented after trampling. Dibben (1971) has successfully used such fragments as innoculum for growing Cladonia species in phytotron experiments. Goniocysts are minute granular thalli (Fig. 4.4) with external monolayered paraplectenchyma and numerous photobiont cells. The formation of goniocysts in the genus Vezdaea may occur by serial de novo lichenization (Scheidegger 1994b) or by ongoing proliferation of a subcuticular thallus (Tschermak-Woess and Poelt 1976) where soredia-like propagules are formed (Ve˘zda 1980; Se´rusiaux 1985). 64 B. Bu¨ del and C. Scheidegger
Table 4.1. Symbiotic and aposymbiotic vegetative propagules
Structure Definition
Goniocyst Granular thalli with external monolayered paraplectenchymatous fungal plectenchyme and numerous photobiont cells Fragment Detached terminal or marginal parts of thallus Phyllidium Corticate, dorsiventral protuberance of upper cortex Bulbil Multilayered paraplectenchymatous globular outgrowth with only a few algal cells Blastidium Pseudocorticate budding proliferations of upper or lower phenocortex Isidium Corticate protuberance of upper cortex and photobiont layer Pseudoisidium Isidia-like structures which lack photobiont cells Polysidium Corticate protuberance of thalline outgrowth Schizidium Flakes of upper cortex and algal layer Soredium Noncorticate clump originating in the medulla and photobiont layer Consoredium Aggregations of incompletely separated soredia Parasoredium Clump of disintegrating (pheno?)cortex and photobiont layer Isidioid soredium Secondarily corticate protuberance produced in soralia-like clusters Dactylidium Phenocorticate nonbudding protuberance of upper cortex and photobiont layer Thlasidium Isidia-like structure internally producing soredia-like granules which are squeezed out Hormocyst Hormogonium with adhering mycobiont Conidium Aposymbiotic, nonmotile, asexual fungal spore; formed within or on a conidioma Thallospore Aposymbiotic, asexual fungal spore produced on the thallus surface or on attachment organs Hormogonium Aposymbiotic, few-celled motile trichomes without heterocysts Zoospore Aposymbiotic, motile daughter cells of the phycobiont
Isidia are scattered across the thallus surface, and their height ranges from around 30 mm to more than 1 mm. Isidia are often cylindrical and simple (Figs. 4.5, 4.23) or branched, but warty or coralloid forms are also known. Although they may serve as diaspores in many species, isidia may also play an important role in increasing thallus surface area. Because numerous isidia are usually present on a thallus surface without being detached, they both increase the photosynthetically active surface (Chapter 9) and enhance other interac- tions with the atmosphere (e.g. trace gas exchange, aerosol deposition; Chapters 11 and 12). Pseudoisidia differ from isidia by the lack of photobiont cells (Walker 1985). Therefore, they would be better treated as fungal (aposymbiotic) propagules, although their function is not fully understood. Thallus morphology and anatomy 65
Phyllidia resemble isidia in very early developmental stages but they soon bend over the parent thallus and become dorsiventral. Usually phyllidia are constricted at the base and become easily detached. In contrast, lobules are similar morphologically, but do not usually act as reproductive propagules. Phyllidia are stratified in a similar manner to the parent thallus (i.e. with upper and lower cortices, a photobiont layer and a medulla). Polysidia are clustered isidia, formed on thalline outgrowths. They have only been reported from Pyxine (Kalb 1987). Thlasidia morphologically resemble pseudoisidia in their terminal ends but contain soredia-like patches with photobiont cells in their bases. Ontogenetically they emerge from the thlasidium. They are only reported from the crustose, epiphytic lichen Gyalideopsis anastomosans (Ve˘zda 1979;Poelt1986). Bulbils are multilayered paraplectenchymatous globular outgrowths of the thallus and have been found in lichenized Basidiomycetes. They contain only a few photobiont cells (Poelt and Obermayer 1990b). Schizidia are formed from upper thalline layers by disintegrating flakes of cortical and algal layers. Schizidia are found in crustose lichens (e.g. the genus Fulgensia) and also occur in foliose lichens, such as Flavoparmelia caperata, Hypogymnia and Xanthoria (Poelt 1980, 1994). Soredia consist of a few photobiont cells enveloped by a loose, spherical mantle of hyphae. Soredia are formed by proliferation of the algal and medullary layers. They are very small and often range from 20 to 50 mm in diameter. Soredia either develop diffusely on the upper surface of the thallus or in delimited areas, called soralia (Figs. 4.6, 4.24). Soralia covered by small soredia are farinose and those with larger soredia are granular. According to their position soralia are classified as laminal, marginal, fissural, cuff-shaped, vaulted, labriform or term- inal. Soredial masses are loosened from soralia and scattered by hygroscopical movements of the cortical tissue (Jahns et al. 1976). Soralia are often hydrophobic, but in repelling raindrops soredia may be removed as well. In many species of Lobaria, Melanelia and other genera, soredia may remain attached in the soralium after their formation. Such soredia may later develop an outer layer of more or less agglutinated hyphae until a cortex or a phenocortex is formed. Such structures are soredia in origin and are called isidioid soredia (Esslinger 1977). Consoredia, a term recently introduced by Tønsberg (1992), refer to aggre- gated soredia that are formed by incomplete division of parent soredia. Parasoredia have a different ontogeny from soredia. They are formed in layers from clumps of disintegrating cortical and photobiont material and are often bigger than soredia (Codogno et al. 1989). Blastidia are pseudocorticate, budding proliferations of the upper or the lower cortex, and sometimes they resemble soredia (Poelt 1980, 1994). 66 B. Bu¨ del and C. Scheidegger
µ µ µ
µ
µ µ
µ Thallus morphology and anatomy 67
The structures described above and summarized in Table 4.1 are often multi- functional organs, for which structure and function may change during their ontogeny. For instance Parmotrema crinitum develops cylindrical isidia (Fig. 4.23) that may serve as diaspores after they become mechanically detached from the thallus. The same structures, however, may develop further on the thallus into multibranched isidia or even into dorsiventral lobules (Ott et al. 1993) that may be interpreted as phyllidia. In Lobaria pulmonaria laminal and marginal lobules develop from isidioid soredia. In the process their function is changed from that of a dispersal propagule to a regenerative structure that enables the re-establishment of a new thallus where the old one grew.
4.4 Evolution of the lichen thallus morphology and resulting functional aspects
Among all symbiotic associations, the lichen symbiosis is unique as the host (mycobiont) expresses its specific phenotype only in association with an adequate photobiont (see also Chapter 5). Thallus morphology is optimized for water uptake and/or loss within a specific habitat, and the anatomical structure of the thallus must follow principles that allow optimal carbon dioxide gain of the photobiont under a given habitat moisture regime.
4.4.1 Morphology and anatomy of fossil lichens Since there is only a very limited record on fossil lichens, we do not know much about the morphology and anatomy of early lichens and how they evolved. The oldest fossil lichen on record is 600 million years old and comes from
Caption for Figs. 4.37–4.43 Functional morphology, fossil record. Figs. 4.37–4.39. Lobaria pulmonaria (Switzerland, corticolous), cross fractures of dry thallus showing cortical and medullary hyphae with cavitations. (LTSEM.) Fig. 4.40. Spongiophyton minutissimum, fossil lichen, vertical section through upper half of the thallus of two neighboring lobes (arrows show lobe margins). (SEM, with permission of International Journal of Plant Science and W. A. Taylor.) Fig. 4.41. Verrucaria elaeina (Germany, saxicolous, submerged), wet thallus of the aquatic lichen, vertical section through perithecium showing asci (as) with spores and the ostiole (os). Paraplectenchymatic thallus structure with airspaces (arrows); sub, substratum. (LTSEM, courtesy of Dr. H. Thues.) Fig. 4.42. Peltula tortuosa (South Africa, saxicolous), cross section through an erect thallus lobe (submerged in water for 6 hours). Medulla (me) still filled with airspaces. (LTSEM.) Fig. 4.43. Peltula tortuosa (South Africa, saxicolous), cross section through thallus margin and medulla (sub- merged in water for 6 hours). Cyanobiont layer and medulla remain free of capillary water, cyanobiont cells turgid (arrow). (LTSEM.) 68 B. Bu¨ del and C. Scheidegger
China (Yuan et al. 2005). In terms of morphology, the most convincing report came from the Lower Devonian Rhynie Cherts (c. 400 million years old) and showed the structure of a gelatinous thallus of the lichen Winfrenatia reticulata, formed by the interaction of nonseptate fungal hyphae surrounding clusters of cyanobiont cells resembling unicellular cyanobacteria of the genera Gloeocapsa or Chroococcidiopsis (Taylor et al. 1997). Another Devonian lichen, Spongiophyton minutissimum, is reported from the Middle Devonian of Brazil and Ghana, and the Lower Devonian of Bolivia and Canada (Taylor et al. 2004). The thallus of this fossil formed by dichotomously to irregularly branching lobes about 1 mm wide between branching points is also of the compact gelatinous type (compare also Collema; Fig. 4.13), with no or only minor airspaces included (Fig. 4.40). From this, one could conclude that the first lichen thalli had cyanobionts and evolved in amphibious habitats where they frequently became submerged. The fact that many modern aquatic lichens, e.g. the genus Verrucaria, still have a compact and sometimes even gelatinous structure (Fig. 4.41) supports that hypothesis. There is a big gap between the Devonian lichen records and the next avail- able fossils. A younger record from tuff rock of the middle Miocene (12–24 million years) can be attributed to the Lobariaceae (Peterson 2000). Another rich source for lichen fossils are amber inclusions which date back as far as 55 to 35 million years and less. However, these lichens very much resemble recent species and already have the morphological features of modern lichens (e.g. Rikkinen 2003).
4.5 Outlook
Although the principal aspects of morphology have been known for many years, functional aspects of lichen morphology are still poorly known. Understanding the functional role of morphology is of major importance to ecophysiology (Chapter 9), reproductive biology (Chapter 6), and systematics (Chapter 17). 5 Morphogenesis
r. honegger
This chapter focuses on the development of the lichen thallus and on key factors playing a role in this fascinating process. The term morphogenesis is derived from Greek, in which ‘‘morph’’ means form and ‘‘genesis’’ means origin or creation.
5.1 Acquisition of a compatible photobiont
Lichen-forming fungi express their symbiotic phenotype (produce thalli with species-specific features) only in association with a compatible photobiont. About 85% of lichen mycobionts are symbiotic with green algae, about 10% with cyanobacteria (‘‘blue-green algae’’) and about 3–4%, the so-called cephalodiate species, simultaneously with both green algae and cyanobacteria (Tschermak- Woess 1988; Persˇoh et al. 2004). Lichen photobionts are extracellularly located within lichen thalli. The enigmatic Geosiphon pyriforme, the only representative of Glomeromycota which does not form arbuscular mycorrhizae with plants (Schu¨ ssler et al. 2001; Schu¨ ssler 2002), is not normally considered a lichen. In this endocyanosis the uptake of cyanobacterial filaments (Nostoc punctiforme) into plasma-membrane-bound vesicles by the fungal protoplast can be readily studied, a feature of considerable interest in cell biology with regard to chloro- plast acquisition (Mollenhauer 1992; Schu¨ ssler et al. 1995; Gehrig et al. 1996; Mollenhauer et al. 1996). A wide range of so-called cyanotrophic lichens form loose, associative sym- bioses with free-living cyanobacteria (Poelt and Mayrhofer 1988), and bacterial films are regularly found on thallus surfaces of lichens (Fig. 5.15). The biology of these associative symbioses has not been experimentally explored. It is highly probable that many of these prokaryotic epibionts are diazotrophic, i.e. capable
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. 70 R. Honegger
of fixing nitrogen (N2) and providing thalli with fixed nitrogen. Other bacterial epibionts might produce hormones or other biologically active compounds. Plant beneficial bacteria and mycorrhization helper bacteria were shown to play important roles in ecosystems (Davison 1988; Garbaye 1994; Haas and Keel 2003; Aspray et al. 2006), but future investigations are needed to establish whether there are also lichen beneficial or lichenization helper bacteria. Only recently was a first molecular analysis on the taxonomic affiliation of lichen- associated bacteria published (Cardinale et al. 2006). The majority of lichen-forming fungi reproduce sexually and thus have to re-establish the symbiotic state at each reproductive cycle. Compatible photobiont cells are not normally dispersed together with ascospores; exceptions are found in a few species of Verrucariales with hymenial photobionts (e.g. Endocarpon pusillum). Many tropical lichen-forming fungi associate with green-algal taxa that are widespread and common in the free-living state (e.g. Cephaleuros, Phycopeltis, Trentepohlia; see Chapter 3), but contradictory views are found in the literature concerning the abundance of representatives of the genus Trebouxia in the free- living state. This is the most common and widespread genus of lichen photo- bionts in temperate regions and especially in extreme climates such as arctic, alpine, antarctic or desert ecosystems. For a long time Trebouxia spp. were postulated to be rare or even missing outside lichen thalli (Ahmadjian 1988, 2001). Based on the assumption of scarcity of free-living photobionts some common and widespread, sexually reproducing Xanthoria species were hypothe- sized to acquire compatible photobionts ‘‘by theft,’’ their ascospore-derived germ tubes entering the symbiotic propagules or thalli of adjacent Physcia spp. and associating with their photobiont (Ott 1987a,b; Ott et al. 2000b). However, in the phycological literature T. arboricola, type species of this genus and photo- biont of innumerable lichen-forming ascomycetes, is common and widespread in the free-living state; the same seems to apply for other Trebouxia species (Ettl and Ga¨rtner 1984; John et al. 2002; Rindi and Guiry 2003). In a series of elegant in situ studies on relichenization processes, carried out on microscopy slides that were incubated in lichen communities, Sanders and Lu¨ cking (2002) and Sanders (2005) showed that free-living Trebouxia cells are abundant and readily available to germ tubes derived from ascospores of lichen mycobionts, Xanthoria spp. included. These authors confirmed earlier reports on the abundance of free-living Trebouxia spp. (Bubrick et al. 1984;Mukhtaret al. 1994). Thus there is no need for ‘‘cleptobiosis’’ in Xanthoria spp., i.e. theft of photobiont cells from adjacent Physcia spp., as postulated by Ott and coworkers (Ott 1987a, b;Ottet al. 2000b). Molecular analyses clearly show that Xanthoria spp. do not associate with photobiont taxa contained in thalli of Physcia spp. of the Physcietum adscendentis (Beck et al. 1998; Dahlkild et al. 2001;Helmset al. 2001; Helms 2003;Nyati2006). Morphogenesis 71
A large number of lichen-forming fungi produce asexual symbiotic propa- gules such as soredia, isidia, or blastidia (see Chapter 4), and dispersal of the symbiotic system via thallus fragmentation appears to be more common than has been previously assumed. The role of vertebrates and invertebrates as vectors of symbiotic propagules, thallus fragments or living cells of lichen- forming fungi and their photobionts remains to be explored. Fecal pellets of the almost ubiquitous lichenivorous mites were shown with culturing experi- ments and molecular techniques to contain viable ascospores and Trebouxia cells (Meier et al. 2002); they seem to be almost ideal propagules for short and long distance dispersal of the symbiotic system. Intact cells of lichen photobionts were also found in fecal pellets of lichenivorous snails (Fro¨ berg et al. 2001).
5.2 Recognition and specificity
About 100 species of cyanobacterial and algal photobionts are known to science. Among the few marine lichen-forming ascomycetes a small number of species associate with Xanthophyceae or Phaeophyceae (Tschermak-Woess 1988; Sanders et al. 2004). Lichen photobionts are less intensely studied than their fungal partners. In less than 2% of lichens has the photobiont ever been identified at the species level; fairly often not even the generic affiliation is known. Many algal or cyanobacterial species are compatible photobionts of large numbers of lichen-forming fungal species from different families; others have only rarely been found in lichen symbiosis. The taxonomic range of acceptable photobionts per lichen-forming fungus can be explored in different ways. Most lichen-forming fungi and their photoautotrophic partners can be isolated into sterile culture and grown in the aposymbiotic state (see the Appendix). Coculturing of fungal and algal isolates would be an elegant mode of studying the range of compatible photobionts per fungal species. Unfortunately the symbiotic phenotype is not routinely expressed in axenic cultures under laboratory conditions, as the factors triggering morphogenesis in compatible combinations are very poorly understood. Due to these problems most investigators explore the range of compatible photobiont species per fungal species by examining the algal or cyanobacterial partner contained in lichen thalli that had been collected in the wild, ideally from a very wide area of distribution. Before the advent of molecular techniques this was a difficult and time-consuming task, which included isolation and sterile culturing of the photobiont under defined conditions and comparison with axenically cultured type strains using light microscopy. Thus only a few experts worldwide were able to identify lichen photobionts at the species level. Today cyanobacterial- or algal-specific primers are used to amplify especially informative photobiont 72 R. Honegger
sequences in DNA extracts of entire thalli. Resulting sequences are compared with sequence data of type strains and others contained in gene data banks. With molecular techniques not only the range of compatible photobionts per fungal taxon can be elucidated, but also the phylogeny of the fungal and photo- autotrophic partners of lichen symbiosis (examples in Kroken and Taylor 2000; Helms 2003; Yahr et al. 2004, 2006). The majority of genera of lichen-forming fungi associate with photobionts of one genus (Rambold et al. 1998). The situation of cephalodiate species is dis- cussed under Section 5.5. Among the noteworthy exceptions are Euopsis grana- tina (Lichinaceae), whose thalli harbor colonies of the cyanobacterium Gloeocapsa sanguinea and of the green alga Trebouxia aggregata beside bacterial colonies (Bu¨ del and Henssen 1988), or Muhria urceolata (Stereocaulaceae), which com- prises areoles with green-algal photobionts overlying a basal hypothallus-like mat containing cyanobacterial filaments (Microcystis and Stigonema spp.; Jørgensen and Jahns 1987). Both examples differ from cephalodiate taxa by not keeping their cyanobacterial partners in confined gall-like structures (see Section 5.5). A remarkable situation was observed in the genus Chaenotheca, whose representatives associate with green-algal photobionts from different genera within the Trebouxiophyceae (Dictyochloropsis, Trebouxia, Stichococcus) and Ulvophyceae (Trentepohlia); however, each Chaenotheca species associates with representatives of only one algal genus (Tibell 2001; Tibell and Beck 2001). Some investigators kept claiming that lichen symbiosis is not specific because sterile cultured lichen-forming fungi tend to overgrow whatever they encounter (Ahmadjian 1988). However, the symbiotic phenotype is expressed only in symbiosis with a compatible photobiont (see Table 5.2). Based on the observation that very common and widespread species of aerophilic algal com- munities are not compatible partners of lichen-forming fungi and on the assumption that Trebouxia spp., the most common lichen photobionts, are rare outside lichen thalli the term selectivity was introduced by Galun and Bubrick (1984). Accordingly, specificity refers to the degree of taxonomic relatedness of compatible partners (Smith and Douglas 1987), and selectivity to their avail- ability in natural ecosystems. In the recent literature both terms are confused (e.g. Rambold et al. 1998). Yahr et al.(2006) define specificity as the range of genetically compatible photobionts; selectivity as the choice of the ecologically optimal genotype among the range of compatible taxa. Data sets from molecu- lar investigations and from field studies, as performed in the last 10 years, significantly improved our knowledge on the availability of photobionts and on the specificity of the lichen symbiosis. However, quite often data sets cannot be compared or combined when investigators focus on different molecular marker sequences. Considerable progress is expected in the coming years. Table 5.1. Specificity and selectivity in morphologically complex macrolichens
Specificity (levels of specificity as summarized by Smith and Douglas, 1987) Very high Acceptable partners found within a subspecific taxon (strain, variety) No known examples in lichens High One species is acceptable. Examples: Parmelia sulcata1 Photobiont: Trebouxia impressa n ¼ 27; Europe, N. America Pleurosticta acetabulum1 Photobiont: Trebouxia arboricola n ¼ 19; central Europe Moderate Acceptable partners found with one genus. Examples: Anzia opuntiella2 Photobionts: Trebouxia gelatinosa (12/39) n ¼ 39; Japan T. potteri (16/39) T. showmanii (11/39) Trebouxia arboricola (3/13) Punctelia subrudecta1 Photobionts: T. gelatinosa (9/13) T. jamesii (1/13) n ¼ 13; central Europe Low to very low Acceptable partners found within the same phylum or within different phyla. Examples: Photosymbiodemes and cephalodiate species: lichen mycobionts (c. 3–4% of lichen-forming fungi) which associate simultaneously or consecutively with selected green-algal and cyanobacterial photobionts (see text for further explanations). Selectivity (see Galun and Bubrick, 1984) Low Large numbers of unrelated taxa are acceptable partners. No known examples in the lichen symbiosis High Few, possibly unrelated, often quite rare taxa are acceptable partners. Examples: (a) Trebouxia spp. (unicellular green algae) are photobionts of > 50% of lichens, rarely found outside lichen thalli. Their mycobionts are moderately to highly specific and highly selective with regard to photobiont acquisition (see above). (b) Lichen mycobionts which form photosymbiodemes or cephalodia associate with very few, selected green-algal and cyanobacterial species, thus revealing low specificity but high selectivity.
Most mycobionts of morphologically complex macrolichens are HIGHLY to MODERATELY SPECIFIC and HIGHLY SELECTIVE with regard to photobiont acquisition. n, Number of thalli investigated. Sources: 1. Friedl (1989b); 2. Ihda et al.(1993). 74 R. Honegger
Presently available data indicate that morphologically advanced taxa of lichen-forming fungi (the ones which form foliose or fruticose thalli with inter- nal stratification) are moderately specific to specific; most green-algal lichens associate with different genotypes of one to a few morphospecies belonging to the same Trebouxia clade or subclade sensu Helms et al.(2001). Examples include various Letharia spp., all symbiotic with genotypes (referred to as phylospecies) of Trebouxia jamesii (Kroken and Taylor 2000), or Physcia, Phaeophyscia, Phaeorrhiza, Physconia and Hyperphyscia spp., all symbiotic with genotypes of T. impressa in subclade I1 (Dahlkild et al. 2001; Helms 2003). Lower specificity was detected in Umbilicaria antarctica and U. kappeni collected along a transect on the Antarctic Peninsula. These morphologically advanced, umbilicate species associated with Trebouxia spp. from two different Trebouxia clades in different regions of the Antarctic (Romeike et al. 2002). In temperate regions, morphologically advanced cyanobacterial lichens select Nostoc spp. from the same monophyletic lineage within the genus; they share some of their photobiont genotypes with repre- sentatives of other cyanobacterial symbioses, such as liverworts, hornworts, or angiosperms (Rikkinen et al. 2002; O’Brien et al. 2005). A lower cyanobiont selectivity was documented in gelatinous and squamulose cyanobacterial lichens and in cephalodiate green-algal lichens in the Antarctic (Wirtz et al. 2003). Some of the morphologically less advanced fungal species, which form crus- tose, nonstratified thalli, were less specific since they associate with photobiont taxa from different clades within the genus Trebouxia sensu stricto. Examples include (1) Rinodina oxydata, which is symbiotic with photobionts of the impressa clade (subclade I5) and the arboricola clade (subclades A6 and A10) (Helms 2003), (2) Lecanora rupicola, which associates with photobionts of the arboricola clade (T. decolorans, T. incrustata and unidentified Trebouxia spp.), the impressa clade (T. impressa), and the simplex clade (T. simplex) (Blaha et al. 2006). Molecular techniques facilitate the study of photobiont diversity within lichen communities (Beck et al. 1998, 2002; Beck 1999; Schaper and Ott 2003); Representatives of the same algal clade, species or even genotype may be con- tained in thalli of unrelated lichen-forming ascomycetes within the community. An example is the Physcietum adscendentis on eutrophicated bark (Beck et al. 1998), in which Trebouxia arboricola of the arboricola clade sensu Helms et al.(2001)is the photobiont of ascospore-producing taxa such as Lecania cyrtella, Lecanora spp., Lecidella elaeochroma (all Lecanoraceae) and Xanthoria parietina (Teloschistaceae). Adjacent sorediate, usually sterile Physcia adscendens and Phaeophyscia orbicularis associate with T. impressa of the impressa clade. With high probability popula- tions of free-living T. arboricola and related taxa from the arboricola clade are present on such substrates and available to ascospore-derived germ tubes of Morphogenesis 75 lichen-forming fungi (Beck et al. 1998; Sanders 2005). Soredia were shown to be wind-dispersed over long distances (Marshall 1996; Tormo et al. 2001; Mun˜oz et al. 2004). Molecular tools facilitate the analysis of genetic variation among mycobionts and photobionts within populations of sexually reproducing (re- lichenizing) and vegetatively propagating lichen species, i.e. sorediate, isidiate taxa, or species dispersing mainly by thallus fragmentation in undisturbed and disturbed collecting sites (examples in Piercey-Normore 2004, 2006; Yahr et al. 2004, 2006; Ohmura et al. 2006). Why are the majority of lichen-forming ascomycetes symbiotic with repre- sentatives of Trebouxiophyceae, especially Trebouxiales? This class of green algae, also referred to as ‘‘lichen algae group’’ (Lewis and McCourt 2004), com- prises also symbionts of protists (Chlorellales) and invertebrates beside numer- ous nonsymbiotic taxa. Chlorella photobionts of freshwater invertebrates and protists differ from nonsymbiotic Chlorella spp. by their continuous release of photosynthates. As shown in resynthesis experiments, this key feature has a signaling effect in the recognition process, but is a serious disadvantage in competitive interactions with nonsymbiotic algae outside the C-heterotrophic exhabitant (Smith and Douglas 1987; Reisser 1992). So far it has not been possible to identify such key features by which lichen photobionts differ from other algae and cyanobacteria. In the symbiotic state, the majority of green-algal lichen photobionts produce and release acyclic polyols such as ribitol, erythri- tol, and sorbitol, depending on their taxonomic affiliation (Smith and Douglas 1987), but it remains unclear whether they do so in the aposymbiotic state in natural ecosystems. Lichen-forming fungi also produce acyclic polyols (mainly mannitol; Smith and Douglas 1987), but different ones from their photobionts. Acyclic polyols serve as compatible solutes in drought-stressed lichen-forming and nonlichenized fungi, plants, algae, archae- and eubacteria (Jennings 1995; Kets et al. 1996; Magan 1997; Martin et al. 1999). It is particularly interesting that both partners of green-algal lichens produce the same group of compounds as osmoregulators. Establishment of a symbiotic relationship is a multistep process, involving preformed compounds and newly synthesized ones. It necessitates the induc- tion and suppression of gene expression in both partners. Only a few studies have explored relichenization events with molecular tools. SEM techniques and cDNA-AFLP were used to investigate gene expression profiles during early re- lichenization of sterile cultured Baeomyces rufus and Elliptochloris bilobata (Trembley et al. 2002c). After one day of coculturing, the photobiont cells were nonspecifi- cally bound to hyphal surfaces by mucilaginous secretions of the mycobiont. After 12 days coculturing the mycobiont started to enwrap the photobiont, and after 28 days the algal cells were enclosed in soredia-like clusters. The analysis of 76 R. Honegger
total mRNA on the twelfth day of resynthesis revealed induction of a few, unidentifiable genes and suppression of many others in both partners (Trembley et al. 2002c). Among potentially preformed substances mediating an initial, often unspecific binding are mycobiont-derived lectins (sugar-specific, cell agglutinating proteins) and other algal-binding proteins (Bubrick and Galun 1980a, b; Hersoug 1983; Kardish et al. 1991; Legaz et al. 2004). One of these unidentified algal-binding proteins was located with immunocytochemical techniques on the wall surface of ascospore-derived germ tubes of Xanthoria parietina (Bubrick et al. 1981). Later developmental stages of morphogenesis in lichens are known from purely descriptive studies only and remain unexplored at the molecular level.
5.3 Morphogenesis
The majority of lichen-forming fungi produce morphologically simple, crustose thalli with no internal stratification (see Chapter 3; Table 5.2). Macrolichens, including foliose and fruticose species with their internally stra- tified thallus, are particularly interesting because they are the morphologically most complex vegetative structures in the fungal kingdom. Such internally stratified lichen thalli are the result of an amazing hyphal polymorphism, which includes either filamentous (polar) or globose (apolar) hyphal growth in combination with hydrophilic or hydrophobic (water-repellent) wall surfaces. Main building blocks of internally stratified lichen thalli are (1) tissue-like areas (pseudoparenchyma) formed by fungal cells which are tightly adhering (conglu- tinate) to each other by means of gelatinous, hydrophilic cell wall material, and (2) loosely interwoven aerial hyphae (plectenchyma) with hydrophobic wall surfaces, some of which are in close contact with the photobiont cells. The gelatinous material (mostly b-glucans) of conglutinate pseudoparenchyma is soft, flexible and translucent when wet but firm, opaque and often quite brittle when dry. Conglutinate pseudoparenchyma in cortical layers and/or in internal strands (Figs. 3.3, 3.14, 5.15, 5.16) provide mechanical stability to the thallus while the system of aerial hyphae with hydrophobic wall surfaces creates gas- filled zones either in the medullary layer in the thalline interior or at the periphery (Figs. 5.14–5.16). As an adaptation to the symbiotic way of life, the thalli of most foliose and fruticose lichens have aerial hyphae in the interior and conglutinate, tissue-like zones at their periphery. The same fungi, when isolated and grown in sterile culture, form conglutinate thallus-like structures with aerial hyphae at their periphery (Fig. 3.3). In the symbiotic phenotype, photo- biont cells are optimally positioned with regard to illumination and gas exchange at the periphery of the gas-filled thalline areas right underneath the Morphogenesis 77
Table 5.2. Lichenization and morphogenesis in morphologically complex macrolichens
Õ germ tubes, free hyphae, rhizomorphs Õ
NON SPECIF IC CONTACT with either Õ compatible aposymbiotic or symbiotica photobiont cells, or with incompatible algae or cyanobacteria Õ Õ
Õ PRETHALLUS stageb: an inconspicuous, nonstratified crust Incompatible associations will not develop beyond this stage unless compatible photobiont cells become available compatible associations only Unknown stimuli, triggering the phenotypic expression (morphotype, chemotype) of the fungal genotype Õ SYMBIOTIC PHENOTYPE: STRATIFIED THALLUS stratification: differentiation of conglutinate zones (usually as peripheral cortex) and gas-filled thalline areas (as internal medulla and algal layer) polarization of the thalline primordium; differentiation of a growing apical and a nongrowing basal thalline area short distance shifting of photobiont cells into and within the algal layer induction of secondary metabolism in the fungal partner, possibly as a response to the establishment of an appropriate nutritional basis. Secondary metabolites crystallize either at the surface of aerial hyphae or within the gelatinous matrix of the cortical layer Õregulation of growth and cell turnover in both partners
MATURE THALLUS: REPRODUCTION AND DISPERSAL
MYCOBIONT: ascospores: sexual reproductive stages conidiac, brood grains: asexual reproduction
SYMBIOTIC SYSTEM (asexual symbiotic propagules): soredia, nondifferentiated isidia internally stratified isidia, phyllidia, thallus fragments a Symbiotic photobionts of either prethallus stages, thalli or symbiotic propagules of vicinal lichen species. b The prethallus stage of development has also been termed: thalle primaire, soredial stage, presquamules, disque primaire, Grundgewebe, basal tissue, preliminary phase. c Macroconidia are seldom produced in macrolichens but have been reported in numerous crustose species, some of which belong to the conidial lichen-forming fungi. Sources: As summarized by Honegger (1993). 78 R. Honegger
conglutinate peripheral cortex. It is interesting to see that these main building blocks are formed by unrelated lichen-forming ascomycetes and basidiomycetes in symbiosis with unrelated green-algal or cyanobacterial photobionts. Wall surface properties of lichen-forming fungi play key roles in thalline water relations and thus in the functioning of the symbiotic relationship. Hydrophilic areas of hyphal walls passively absorb and retain water, but hydro- phobic wall surface layers, as typically found on aerial hyphae of lichenized and nonlichenized fungi, prevent free water from accumulating; thus the medullary and algal layers of lichen thalli stay gas-filled at any level of hydration, and this is a prerequisite for efficient gas exchange (Honegger 1997, 2001, 2006). Already Goebel (1926) realized that medullary hyphae of internally stratified lichen thalli have water-repellent wall surfaces, which he assumed to result from depositions of mycobiont-derived, crystalline secondary metabolites. However, wall surface hydrophobicity is also evident in the thalline interior of lichens with no medullary secondary compounds; examples are Xanthoria and Peltigera species. In freeze-fracturing preparations, a thin, proteinaceous wall surface layer was observed not only on hyphae, but also on algal wall surfaces in the medullary and algal layers of various taxa, which revealed a peculiarly fine structure: groups of minute rodlets, lying in parallel (Fig. 5.15, lower panel), create a hydrophobic discontinuity (Honegger 1982b, 1984a; Scherrer et al. 2000; Trembley et al. 2002a). The same wall surface layer, termed rodlet layer, was found on aerial hyphae of numerous nonlichenized fungi. In biochemical and molecular studies the rodlet layers of nonlichenized and lichenized ascomycetes and basidiomycetes were identified as hydrophobins, a class of fungal surfac- tants with very peculiar properties (Kershaw and Talbot 1998; Wessels 1999; Scherrer et al. 2000, 2002;Wo¨ sten 2001; Trembley et al. 2002a, b; Scherrer and Honegger 2003). Hydrophobins are small, secreted fungal proteins (approx. 100 amino acids long), which self-assemble at liquid–air or hydrophilic–hydrophobic interfaces to a thin, amphiphilic film with a distinct rodlet (i.e. small rods) structure on its hydrophobic side (Fig. 5.15, lower panel; Wo¨ sten et al. 1993; Wessels 1999;Wo¨ sten 2001; Linder et al. 2005). Once assembled, a hydrophobin film cannot be dissolved with techniques that are usually applied for protein solubilization. Hydrophobins reveal very low sequence homology except eight cysteine residues in a conserved pattern (Fig. 5.15, lower panel); these form intramolecular disulfide bonds (Kershaw et al. 2005; Linder et al. 2005). Many fungal species form several hydrophobins, which are all differerent from each other. Only one hydrophobin was found in each of different Xanthoria spp. (Scherrer et al. 2000; Scherrer and Honegger 2003), but three were found in the dikaryotic hyphae of the lichenized basidocarps of Dictyonema glabratum; these share between 54% and 66% amino acid identity (Trembley et al. 2002a). Morphogenesis 79
Hydrophobins play important roles in the establishment of fungal interac- tions with plants and animals (Whiteford and Spanu 2002; Kershaw et al. 2005; Linder et al. 2005). In the lichen symbiosis, they seal the apoplastic continuum of both partners with a hydrophobic coat, which prevents free water from accu- mulating on these hydrophobic wall surfaces and forces the passive fluxes of solutes from the thalline surface to the algal layer and vice versa to flow under- neath this hydrophobic lining (Honegger 1985, 1991a, c, 1997, 2001). As in nonlichenized fungi, hydrophobin gene expression is developmentally regu- lated in lichen-forming ascomycetes and basidiomycetes. XPH1 is expressed in the aerial hyphae of the medullary and algal layers, but not in pycnidia or in the hymenial and adjacent subhymenial layers of Xanthoria parietina (Scherrer et al. 2002). DGH1 and DGH2 are expressed in aerial hyphae of the photobiont layer and DGH3 in hyphae of the boundary layers in the lichenized basidiocarps of Dictyonema glabratum (Trembley et al. 2002b). Due to their low sequence homology even within closely related species, hydrophobins would be excel- lent molecular markers in phylogenetic analyses (Scherrer and Honegger 2003), but they are often too difficult to characterize. There are no conserved regions for which primers could be designed to amplify hydrophobin gene sequences. Thin hydrophobic surface layers often overlie very thick hydrophilic areas of the fungal cell wall. For example, the very thick walled medullary hyphae of Xanthoria spp. form hydrophilic, conglutinate strands with hydrophobin- derived hydrophobic surface coats, and the very thick walled aerial hyphae of the medullary layer in Cetraria islandica (‘‘Icelandic moss’’) comprise a massive, hydrophilic layer composed mainly of lichenin (a linear (1!3) (1!4) b-glucan), which overlies the cell wall proper (Honegger and Haisch 2001); both absorb and retain water. Such hydrophilic wall layers are thick in the hydrated state and thin in the desiccated state. The subsequent developmental steps in the ontogeny of macrolichens are summarized in Table 5.2. Starting with either free hyphae or symbiotic propa- gules many mycobionts first produce a prethallus, i.e. an inconspicuous crus- tose structure with no internal differentiation (Fig. 5.4). These prethallus stages may be formed even with ultimately incompatible algal cells (Ahmadjian and Jacobs 1981; Ott 1987a, b; Schaper and Ott 2003). The most fascinating event in lichen morphogenesis is the least understood: the onset of thallus formation (Figs. 5.5–5.10). How does it become morphologically and physiologically differ- entiated from the prethallus stage? The factors that trigger the morphotypic and chemotypic expression of the fungal genotype are unknown, but the series of events is impressive (Table 5.2). Even the earliest developmental stages show a distinct polarity. The growing apical and the nongrowing basal poles are 80 R. Honegger
1 2
3 4
Figs. 5.1–5.4 Thallus ontogeny in the foliose macrolichen Xanthoria parietina. Fig. 5.1. Prethallus stage in an axenic, artificial combination of a cultured multispore isolate of the mycobiont and an isolate of the compatible photobiont, Trebouxia arboricola. Fig. 5.2. A germ tube of the mycobiont in search of compatible photobiont cells in a natural association of aerophilic green algae on bark. The fungal hypha is seen in close contact with cells of Desmococcus sp. (spiny cell wall surface), an incom- patible green alga. Fig. 5.3. Mycobiont hyphae in a natural resynthesis, secreting mucilaginous material in initial response to the contact with compatible algal cells. Fig. 5.4. Prethallus stages of development in nature (Note the different magnifications in Figs. 5.3 and 5.4!). Magnification bars represent 10 mm. Morphogenesis 81
5 6 20 μm
20 μm
7 50 μm 8 50 μm
9
20 μm
Figs. 5.5–5.9 Thallus ontogeny in the foliose macrolichen Xanthoria parietina (continued). Figs. 5.5, 5.6. External view and longitudinal cross section of primordial stages of the symbiotic phenotype growing out of the prethallus stage of development. These minute primordia are polarized (growing/nongrowing poles being defined) and recognizable by their bright yellow, cortical anthraquinones. The arrow in Fig. 5.6 points to the direction of growth. Figs. 5.7–5.9. External view and longitudinal cross section of juvenile thallus lobules. An internal stratification (upper and lower cortical layers and algal layer) is obvious. The medullary layer (see Fig. 3.3) will be differen- tiated in a later developmental stage. 82 R. Honegger
differentiated (Figs. 5.5, 5.6). Such minute primordial stages reveal an internal stratification: the algal cells are kept in the gas-filled internal zone which is surrounded by a peripheral conglutinate cortex (Figs. 5.6, 5.8, 5.9). The medul- lary layer (Fig. 3.3) is not yet fully differentiated. Species-specific secondary metabolites are already produced; the example shown in Figs. 5.5 and 5.6 are bright yellow owing to the presence of the cortical anthraquinone parietin. Neighboring, potentially genetically heterogeneous prethalli and/or primordia may fuse (Fig. 5.10). The resulting thallus rosette may therefore be genetically heterogeneous.
5.4 Structural and functional aspects of the mycobiont–photobiont interface
Depending on the taxonomic identity of the partners, a variety of appressorial and haustorial structures is found at the immediate mycobiont– photobiont interface (Honegger 1991a, 1992). The structure and composition of the algal cell wall is centrally important to the outcome of the symbiotic relationship. Algal walls with enzymatically nondegradable biopolymers (sporopollenin-like compounds, as found in the genera Coccomyxa and Elliptochloris) are not normally invaded by fungal haustoria (Brunner and Honegger 1985; Honegger 1991a). Correlations between thallus morphology and type of mycobiont–photobiont relationship are obvious in Lecanorales with photo- bionts of the genera Trebouxia and Asterochloris (Tschermak 1941; Plessl 1963; Honegger 1986a, 1992, 2001). Mycobionts of morphologically simple crustose taxa often pierce the algal cell wall with finger-shaped protrusions (‘‘intracel- lular’’ haustoria; Fig. 5.11), that are in intimate contact with the invaginating plasma membrane of the photobiont. Mycobionts of morphologically advanced, foliose or fruticose Lecanorales and Teloschistales form very peculiar intrapar- ietal haustoria (intra: within; paries: wall) that enter, but do not penetrate the cellulosic wall of Trebouxia photobionts (Figs. 5.13, 5.15). Intermediate types of interactions occur where the fungal partner forms a short, almost globose protrusion that is enveloped by the locally enlarged algal wall (Fig. 5.12). Such intimate contact sites are established between juvenile cells: growing hyphal tips of the fungal partner meet young algal cells when these are still ensheathed by the degrading mother cell wall (Fig. 5.15). Subsequently, the haustorial complex on the fungal side and the algal cell grow and develop coordinately to reach maturity. At the first contact of the growing hyphal tip with the wall surface of juvenile photobiont cells, the thin, proteinaceous, water-repellent cell wall surface layer (which was identified as hydrophobin in some species; see above) Morphogenesis 83
10
Fig. 5.10 Thallus ontogeny in the foliose macrolichen Xanthoria parietina (continued). Vicinal, juvenile thallus lobules, growing out of different prethalli, fuse to form a rosette-like thallus which is likely to be genetically inhomogenous. Arrows point to further prethallus stages. of the mycobiont spreads over the algal cell wall surface, thus sealing the apo- plastic continuum of both partners with a hydrophobic coat (Honegger 1984a, 1986b, 1991a, 2001). In many taxa, especially in Lecanorales, the hydrophobicity of this mycobiont-derived, water-repellent wall surface is increased by secondary metabolites that crystallize on and within this water-repellent surface layer (Fig. 5.14;Honegger,1986b, 2001, 2006). Water and dissolved nutrients and soluble metabolites of fungal and algal origin are passively translocated within the apoplastic continuum underneath this hydrophobic coat. The regularly occur- ring, often quite dramatic wetting and drying cycles are the main driving forces in more distant translocation processes (Honegger 1991a, 1992). 84 R. Honegger
P P ic
M
ip 11 12
P P
lp ip
13 14
Figs. 5.11–5.14 Mycobiont (M) – photobiont (P) relationships in Lecanorales with Trebouxia spp. Fig. 5.11. Finger-like, intracellular fungal haustoria (ic) are typically found in photobiont cells of the crustose Lecanora conizaeoides. Fig. 5.12. Almost globose, intraparietal haustorial structures (ip) are produced at the mycobiont–photobiont interface in Cladonia arbuscula. The algal cell wall is not pierced, but grows around the haustorial complex. The arrow points to the border between algal and fungal cell walls. Fig. 5.13. Intraparietal haustorium (ip) at the mycobiont–photobiont interface in Alectoria ochroleuca. This haustorial type is found in foliose and fruticose lecanoralean macrolichens (except Cladoniaceae). Fig. 5.14. Low temperature scanning electron micrograph of the algal layer in a water-saturated, frozen-hydrated sample of Parmelia sulcata. Mycobiont and photobiont cells are coated by a mycobiont-derived, hydrophobic wall surface layer. Secondary metabolites of fungal origin (mainly salazinic acid) crystallize in and on this water-repellent coat, thus increasing its hydrophobicity. lp: crystalline lichen products (mycobiont-derived secondary metabolites). (For further details see Honegger, 1986a, b, 1991a). Magnification bars represent 5 mm. Morphogenesis 85
b b
rcc
cc lp
py
ah ∗ n
ip
n ip
asp
mcw ∗
20 μm
Hydrophobins: small, secreted proteins of lichen-forming and non-lichenized fungi
0.1 μm X 26-85 -C-X 5-8 -C-C-X 17-39 -C-X 8-23 -C-X 5-6-C-C-X 6-8 -C-X 2-13
Fig. 5.15 (Upper panel) TEM micrograph of a vertical cross section of the congluti- nate upper cortex (uc) and algal layer in the growing, marginal zone of the foliose macrolichen Parmelia tiliacea (photobiont: Trebouxia impressa). ah, aerial hyphae of the thalline interior. Their cell wall surfaces were originally covered with crystalline, 86 R. Honegger
5.5 Genotypes and phenotypes
One of the most fascinating aspects of lichen biology is the impact of the photobiont on the expression of the symbiotic phenotype in the fungal partner. Complex morphotypes are expressed exclusively in compatible asso- ciations (Table 5.1). Particularly interesting are the 3–4% of lichen-forming fungi that associate simultaneously (cephalodiate species) or consecutively (photosymbiodemes) with both green-algal and cyanobacterial photobionts. Cephalodiate lichen mycobionts have a green alga as primary photobiont and additionally incorporate nitrogen-fixing cyanobacteria in gall-like structures, the cephalodia. These are located either on the thallus surface (external cepha- lodia; Fig. 5.16) or in its interior (internal cephalodia). The fungal partner forms a dense, conglutinate cortical layer around cephalodia, thus creating microaerobic conditions in its interior. Cephalodia-bound cyanobacteria often reveal an increased heterocyst frequency and thus an elevated nitrogenase activity compared with the aposymbiotic (‘‘free-living’’) state (Englund 1977). Most cephalodia are morphologically simple globose, sacculate, lobate or cor- alloid structures that differ from the rest of the vegetative thallus by their coloration. But even morphologically simple cephalodia illustrate a biologi- cally most fascinating phenomenon: the different response of the mycobiont to a selected cyanobacterial partner compared with the green-algal photobiont (Fig. 5.16). Mycobionts of photomorph pairs (also termed morphotype pairs, photosym- biodemes, etc.) form either a cephalodiate green-algal morphotype (usually vivid green with dark cephalodia) or a cyanobacterial morphotype (usually grayish to black). Both morphotypes of the same fungal species derive their coloration
Caption for Fig. 5.15 (cont.) secondary metabolites (see Fig. 5.14). asp, autosporangium of the photobiont; the thin-walled, young autospores are ensheathed by the degrading mother cell wall (mcw). Arrows point to adjacent mycobiont hyphae entering the autosporangium in order to establish contact sites with the juvenile algal cells. b, bacterial epibionts; cc, cortical cells; ip, intraparietal haustoria entering, but not piercing, the algal cell wall (see Fig. 5.13). lp, impressions of crystalline lichen products (mainly atranorin) in the conglutinate upper cortex. The crystals were dissolved during dehydration. n, nucleus of the algal cell; py, pyrenoid of the large, lobate chloroplast; rcc, remains of decaying cortical cells. Asterisks mark the gas-filled space in the thalline interior. (Lower panel) Rodlet layer, built up by self-assembled hydrophobin protein on the wall surface of aerial hyphae in the medullary and algal layers of asco- and basidiomycetes. See text for further explanations. Morphogenesis 87
Fig. 5.16. Diagram illustrating the impact of the photobiont on the expression of the symbiotic phenotype in the mycobiont as observed in the cephalodiate lichen Stereocaulon ramulosum. Line drawings by Sibylle Erni.
from their photobiont. Photomorph pairs may be morphologically similar (iso- morphic), as seen in various Peltigera spp. (Brodo and Richardson 1978; Tønsberg and Holtan-Hartwig 1983; Goffinet and Bayer 1997) or dissimilar (hetero- morphic) as observed in Sticta/Dendriscocaulon, Pseudocyphellaria/Dendriscocaulon or Lobaria/Dendriscocaulon photomorph pairs (reviewed in James and Henssen 1976). Heteromorphic photomorph pairs comprise a foliose, dorsiventrally organized green-algal (chloromorph) and a shrubby (small fruticose) cyanobac- terial phenotype (cyanomorph), both being described under different genera and species names in the lichenological literature. Intermediate and intermixed growth forms were described as ‘‘lichen chimerae’’: green, foliose thalli with large, blackish shrubby outgrowths resembling outsized cephalodia, or shrubby, blackish cyanomorphs with green, dorsiventrally organized ‘‘leaflets.’’ Molecular analyses confirmed the taxonomic identity (conspecificity) of Peltigera spp. and Lobaria spp. involved in photomorph pairs (Goffinet and Bayer 1997; Stenroos et al. 2003). Thus photomorph pairs mirror an astonishing pheno- typic plasticity, which allows a few lichen-forming ascomycetes to express different phenotypes in association with either a green-algal or a cyanobacterial 88 R. Honegger
partner. Contrasting with this plasticity are the very rigid nomenclatural rules (Heidmarsson et al. 1997; Jørgensen 1998). How should fungi of photo- morph pairs, which are known under two different species or even genus names, be adequately named? Which one of the old names should be given priority? The question of whether the same fungus would differentiate morphologically different symbiotic phenotypes with different green-algal partners was repeat- edly asked. It is theoretically imaginable that morphologically similar pheno- types, which are described under different species names, might be produced by one fungus in association with different green-algal partners. However, until now no such cases are known.
5.6 Growth
In biology textbooks lichens are often referred to as extremely slow- growing and long-lived organisms. Lichens of extreme climates such as deserts and arctic/alpine or antarctic ecosystems often have only short periods during which full metabolic activity and growth can occur (Kappen 1988, 1993). Consequently, only very low cell turnover rates and minimal annual size increases are recorded. However, as shown in a 20-year survey of antarctic lichen communities, not all lichen species from extreme climates are extreme slow-growers (Lewis Smith 1995). Some lichens are assumed to be very long- lived. If true, then their growing thalline areas must be at least minimally active for decades or centuries. The most extreme age estimates are in the range of millennia, as concluded from annual measurements of different size classes of thalli. However, in very large thalli, one can never exclude the possibility that previous fusion of neighboring thalli has occurred. Many lichen-forming fungi do not form a distinct rim due to vegetative incompatibility, when bordering upon other thalli of the same species. Instead they fuse and consequently appear as one entity. In contrast, short-lived lichen species terminate their full devel- opment within months or a few years (Poelt and Veˇzda 1990). The majority of lichens of temperate or subtropic to tropic climates have annual radial growth (or annual linear elongation in the case of fruticose species) in the range of millimeters to a few centimeters. The highest growth rates are recorded in moist, coastal-influenced regions, including such species as Ramalina menziesi, thei ‘‘lace lichen’’ (Figs. 5.24–5.25; Boucher and Nash 1990 a) and lungworts like Lobaria oregana (Rhoades 1977) and L. pulmonaria in rain forests of the Pacific Northwest of North America. Under favourable climatic conditions these taxa reveal growth patterns that allow rapid size increase in relatively short periods of time (Table 5.3). Morphogenesis 89
17 18 19
ac
ac
u
20 21 22
ac
so
u
Figs. 5.17–5.22 Growth patterns in macrolichens. Bars represent 2 mm. Fig. 5.17 Caloplaca thallincola, Fig. 5.18 Teloschistes chrysopthalmus, and Fig. 5.20 Menegazzia pertusa: thalli with predominantly apical/marginal growth, youngest portions at the marginal pseudomeristems and ascomata (ac) or soredia (so) being produced in nongrowing subapical areas. Figs. 5.19, 5.22. Lasallia pustulata, upper and lower surface: irregular, patchy intercalary growth, some of the oldest thalline areas are located at the fuzzy, isidiate margins. u, central umbilicus (holdfast) by which the whole thallus is fixed to the substratum. Fig. 5.21. Peltigera venosa, lower surface: predominantly apical/marginal growth, ascomata (ac) are produced in apical thalline areas which terminate growth once that reproductive stages are formed. 90 R. Honegger
Table 5.3. Growth patterns in foliose and fruticose macrolichens with internally stratified thallus (as summarized by Honegger, 1993)
(1) Predominantly apical/marginal growth Laminal size increase is accomplished by an apical/marginal pseudomeristem and the adjacent elongation zone; highest cell turnover rates occur in the apical/marginal pseudomeristem. Limited cell turnover and fully differentiated mycobiont and photobiont cells are typically found in the nongrowing subapical area. Increasing numbers of dead photobiont cells occur in senescent basal thalline areas. Central senescent parts of foliose thalli may break off, thus giving the thallus a ring-shaped outline. Producton of sexual or (symbiotic) asexual propagules (soredia, isidia) either – in apical/marginal pseudomeristematic zones of lobes or branches which terminate growth once that reproductive stages are formed. examples: Peltigera venosa and other Peltigera spp. Cetraria islandica (‘‘Iceland moss’’) Teloschistes chrysophthalmus – in subsenescent, nongrowing thalline areas; examples: Caloplaca thallincola and other Caloplaca spp. Xanthoria parietina Menegazzia pertusa (2) Combined apical/marginal and intercalary growth, best visible in reticulate thalli with increasing mesh size towards the basal/central parts in either erect or pendulous fruticose species and in foliose lichens which adhere to the substratum with only a small central part of the thallus. Laminal size increase due to the activities of an apical/marginal pseudomeristem and intercalary growth processes. examples: Lobaria oregana, L. pulmonaria (‘‘lungwort’’) and other Lobaria spp. Ramalina menziesii (‘‘lace lichen’’) (3) Regular or irregular (‘‘patchy’’) intercalary growth in foliose, umbilicate thalli of the Umbilicariaceae; best visible in pustulate species. These thalli typically show irregular intrathalline gradients of photosynthetic activity, often highest at the central umbilicus. Some of the oldest thalline areas are likely to be at the fuzzy margins. examples: Lasallia pustulata Umbilicaria spp. Combinations of (1), (2) and/or (3) are likely to occur.
Longevity in lichens needs to be critically interpreted. Pseudomeristems of macrolichens (see below) persist, as do meristems of perennial plants. However, in very old trees the metabolically active cells are not centuries old, and the same is true of old thalli of morphologically complex lichens. Senescent thalline areas are either overgrown or disintegrate. In the case of foliose taxa, the Morphogenesis 91
23 25
24
Figs. 5.23–5.25 Growth patterns in macrolichens (continued): Ramalina menziesii (‘‘lace lichen’’) with combined apical/marginal and intercalary growth. When apical pseudomeristematic zones (Fig. 5.23) are lost by damage or arthropod grazing (Fig. 5.24) the band-shaped, perforate thallus lobe will grow by intercalary elongation, visible in the increasing length of the meshes (Fig. 5.25) (for further details see Sanders, 1989). Magnification bars represent 0.1 mm. remaining thallus becomes strikingly ring-shaped. In very cold or very hot cli- mates it may take centuries to achieve a wide ring diameter, but only the young- est portions, i.e. the growing marginal area and adjacent zones, are retained.
5.6.1 Growth patterns Mycobionts of the morphologically less advanced crustose lichens grow more or less like molds: either on or within the substrate where they meet their compatible photobiont (see Chapter 3). The situation is very different in squa- mulose, foliose, or fruticose lichens with an internally stratified thallus. The coordinated growth of the dominant fungal exhabitant (the partner that lives outside) around the photoautotrophic inhabitant (the photobiont) is a most remarkable biological phenomenon. Three distinct growth patterns have been recognized in macrolichens (Table 5.3, Figs. 5.17–5.25) and intermediates are likely to occur. Foliose and fruticose lichens are more extensively investigated. 92 R. Honegger
New cells are produced almost exclusively in terminal or marginal thalline areas with meristematic properties. Because these growing edges differ in many respects from meristems of plants (Fletcher 2002) and, moreover, because fungi have no real tissues, these growing edges or tips have been termed pseudomer- istems (Honegger 1993). High cell turnover rates and small average cell sizes in both partners are typical features of such pseudomeristematic zones (Fig. 5.26a). Both mycobiont and photobiont cells achieve their full size in the elongation zone behind the pseudomeristem (Fig. 5.26b, c). Fully differentiated fungal and photobiont cells have low cell turnover rates (few or no cell divisions). A high
Xanthoria parietina (Teloschistales, Ascom ycotina) photobiont: Trebouxia arboricola (Chlorophyta)
a foliose, dorsiv entrally organized macrolichen with mainly apical/marginal and limited intercalary growth
pseudomeristematic subapical elongation zone fully differentiated marginal rim thalline area
(a) (b) (c) (d )
uc
ph
m
lc
50 μm*
0.1 mm 1.5 mm 15 mm behind the margin
Mycobiont and photobiont: Mycobiont and photobiont: high cell turnover rates increasing cell size, decreasing cell turnover rates Photobiont: increasing number of oversized cells**
* same magnification in (a) – (d) ** cells having exceeded the size required f or autospore formation without undergoing mitosis, i.e . with arrested cell cycle (Hill, 1985, 1989). Arrow in (a) points to an auto- sporangium. lc: lower cortex; uc: upper cortex; m: medullary layer; ph: photobiont layer;
Fig. 5.26 Internal thalline differentiation in Xanthoria parietina as seen in differential interference contrast light microscopy of semithin sections. Morphogenesis 93 percentage of oversized photobiont cells (cells having exceeded the size required for autospore production without undergoing mitosis; Hill 1985, 1989; Fiechter 1990) are typically found in adult thalline areas (Fig. 5.26d; Honegger 1993). A large number of lichens do not grow exclusively at their margins or tips and adjacent elongation zones (Fig. 5.26a–c) but retain the capacity to expand and enlarge even in their older parts (Chapter 2). Such intercalary (inserted) growth processes have not yet been analyzed in detail, and it is unknown whether they are due primarily to elongation of cells that have been produced by the apical/ marginal pseudomeristem or to continuous cell division. The growth pattern of umbilicate lichens, which are fixed to the substrate by means of a central umbi- licus (i.e. navel; Figs. 5.19, 5.22), are least understood. Umbilicariaceae have no distinct marginal pseudomeristem but reveal irregular, patchy intercalary growth (Honegger 1993, 2001;Figs.5.19, 5.22). Patterns of productivity (photosynthetic activity) are also diffuse in such thalli (Larson 1983;Hestmarket al. 1997). The regulatory mechanisms behind all these differentiation processes in morpholo- gically complex macrolichens are poorly understood. It is inferred from presently available data that the cell turnover in the photobiont is strictly controlled by the mycobiont. The cell cycle of the photobiont in nongrowing thalline areas is arrested (Hill 1985;Honegger1993). As inhibitors of cell division are of general biological interest, especially with regard to tumor suppression in humans and vertebrates, the molecular basis of this inhibitory principle in mycobiont–photo- biont relations of lichens merits thorough investigation. 6 Sexual reproduction in lichen-forming ascomycetes
r. honegger and s. scherrer
A high percentage of lichen-forming ascomycetes reproduce sexually and thus are assumed to disperse primarily via ascospores, which have to relichenize. However, one should keep in mind that even fertile lichens have options for vegetative dispersal in the symbiotic state, either via symbiotic propagules such as soredia, blastidia or isidia, or via thallus fragmentation. Viable fungal and algal cells were shown to be contained in fecal pellets of lichenivorous slugs (McCarthy and Healey 1978; Fro¨ berg et al. 2001) and of the ever-present lichenivorous mites (Meier et al. 2002). Thus, it is not known how often relichenization occurs in natural habitats. A detailed knowledge of sexual reproductive strategies is required for under- standing evolutionary traits and population genetics. Zoller et al.(1999) were the first to recognize that lack of ascomata in strongly fragmented and geographi- cally isolated populations of Lobaria pulmonaria (‘‘lungwort’’) might be due to missing mating partners. As this species produces abundant isidiate soredia, one might conclude that ascospores are unnecessary. However, recombination as the centrally important element of sexual reproduction has an impact on genetic stability, whereas favorable and unfavorable mutations are transmitted to the offspring in clonal (vegetative) dispersal. As pointed out by Hestmark (1992), sexual reproduction may often be a mode of escape from old, severely parasitized thalli (Seymour et al. 2005b). It remains to be seen how often new thalli are formed from germinating ascospores in rarely fertile species with efficient dispersal via vegetative symbiotic propagules, such as Pseudevernia furfuracea, Hypogymnia physodes and others. The majority of lichen-forming ascomycetes can be cultured in the aposym- biotic state, but they fail to differentiate sexual reproductive stages under these conditions; thus, classical genetic crossing experiments cannot be conducted.
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. Sexual reproduction in lichen-forming ascomycetes 95
This problem occurs also in many groups of nonlichenized fungi. In most lichens the vegetative mycelium is not hidden in the substrate or host, as is the case in nonlichenized taxa, but fully visible above ground. Therefore, the abundance of ascomata and their locationonthevegetativemyceliumcanbe evaluated. Some species of lichen-forming ascomycetes are always fertile and have many ascomata (e.g. Xanthoria parietina [Fig. 6.1], Ramalina fastigiata), others have no or few to many ascomata (e.g. Xanthoria calcicola [Fig. 6.4], Parmelia tiliacea, P. sulcata), and in a third group ascomata are very rare (e.g. Pseudevernia furfuracea, Hypogymnia physodes)orabsent(Thamnolia vermicularis, Lepraria, and Leprocaulon spp.). In some species ascomata are formed in the subapical part of the thallus, which continues growth at its periphery and may reach large dimensions (Fig. 5.17); examples are Xanthoria parietina, Parmelia tiliacea, Ramalina menziesii, R. fraxinea, Lobaria pulmonaria.Otherspro- duce ascomata at or near the growing tip or margin which subsequently stops growth; thus, only a limited size can be achieved. Examples are Teloschistes chrysopthalmus (Fig. 5.18) and many other Teloschistes spp., Ramalina fastigiata, and Peltigera venosa (Fig. 5.21).
6.1 Mating systems
Nonlichenized and lichen-forming ascomycetes are not female or male; each haploid mycelium is theoretically capable of differentiating both game- tangia (ascogonia) and gametes (microconidia ¼ spermatia). Their sexual repro- duction is regulated by mating type (MAT) genes (review: Debuchy and Turgeon 2006). In contrast, basidiomycetes have several MAT loci and thus very complex mating systems. Filamentous ascomycetes have one MAT locus, which is com- pletely different in haploid mycelia carrying only one out of the two MAT alleles of the same heterothallic (cross-fertilized) species. They are referred to as MAT 1-1 and MAT 1-2 (Turgeon and Yoder 2000), but other terms are found in the literature as well (e.g. MAT 1 and MAT 2, MAT A and MAT a, etc.). As MAT 1-1 and MAT 1-2 are completely different within the same species the term idiomorph is used instead of allele, one idiomorph carrying one to several genes. Homothallic (self-fertile) species have either elements of both MAT idiomorphs in one haploid mycelium, or of only one, the other being lost. Each haploid mycelium forms ascogonia, but in cross-fertilized species a dikaryon can only be formed with a mating partner carrying the complementary idiomorph. Self-fertile species do not need a mating partner. Homothallism is a derived character (Yun et al. 1999), which can be achieved in only one mutation (Po¨ ggeler 1999). MAT genes evolve very rapidly, but need to be conserved within a species. Thus, mutations in MAT idiomorphs might play important roles in speciation (see below). 96 R. Honegger and S. Scherrer
How can mating systems (homothallism or heterothallism) of lichen-forming ascomycetes be explored when sterile cultured mycelia do not form sexual reproductive stages? The progeny of meiosis can be analyzed with fingerprint- ing techniques. Genomic DNA derived from single ascospore isolates from one ascoma (Murtagh et al. 2000; Seymour et al. 2005a) or from one ascus (Honegger et al. 2004) can be subjected to RAPD-PCR (randomly amplified polymorphic DNA-polymerase chain reaction) or AFLP (amplified fragment length poly- morphisms) and resulting products be screened for polymorphisms. In homo- thallic species, all sporelings from the same ascus and ascoma, and sterile cultured vegetative mycelium of the mother thallus, reveal identical finger- prints (Fig. 6.3). Due to recombination events, polymorphisms are found among RAPD or AFLP markers of sibling isolates in heterothallic species (Fig. 6.5; Murtagh et al. 2000; Honegger et al. 2004; Seymour et al. 2005a; Honegger and Zippler 2007). Already the sporeling phenotype may provide interesting information about the mating systems. In heterothallic species of numerous Teloschistaceae, Parmeliaceae and Physciaceae, all with 8-spored asci, a maximum of four distinct sporeling phenotypes per ascus were observed, which differed in growth rate, growth pattern and/or pigmentation (Fig. 6.5), but in the homothallic Xanthoria parietina (Fig. 6.2) and X. elegans all sporelings per ascus and ascoma grew equally fast and looked the same (Honegger et al. 2004; R. Honegger, unpublished). Heterothallism is easily detectable with fingerprinting techniques, but homothallism is more difficult to identify. Uniform fingerprints among sibling single spore isolates obtained with large numbers of markers (PCR with 10–30 primers, Murtagh et al. 2000; Honegger et al. 2004) suggest homothallism, but the situation can only be properly interpreted when MAT genes are characterized. As MAT genes evolve rapidly and reveal little sequence homology they are difficult to track down, especially when the genome of the species in question has not yet been fully sequenced (as is so far the case in all lichen-forming ascomycetes). Successful characterization of MAT genes in a range of Xanthoria spp. confirmed heterothallism in Xanthoria polycarpa and X. flammea, but homothallism in Xanthoria parietina, with MAT 1-2 present in all sibling isolates, and in X. elegans, with MAT 1-1 and MAT 1-2 in all sibling isolates (Scherrer et al. 2005). MAT 1-2 of Cladonia galindezii was characterized and identi- fied in 40–60% of randomly selected siblings, a further proof of heterothallism in this species (Seymour et al. 2005a). Does the abundance of ascomata per thallus correlate with mating systems? Most investigators wish to know more about mating systems in lichen-forming ascomycetes without investing in time-consuming laboratory experiments. In Teloschistaceae, Parmeliaceae, Ramalinaceae and Physciaceae all irregularly fertile species with no or few to many ascomata turned out to be heterothallic, Sexual reproduction in lichen-forming ascomycetes 97
Figs. 6.1–6.6 Mating systems in lichen-forming ascomycetes visualized with RAPD- PCR fingerprinting techniques applied to genomic DNA derived from single asco- spore isolates. Figs. 6.1–6.3. Thalli of the self-fertile (homothallic) Xanthoria parietina (yellow wall lichen; Fig. 6.1) are always richly fertile, their older parts being covered by apothecial disks (fruiting bodies). All eight ascospores derived from one ascus were manually separated and allowed to grow as single spore isolates (ss). Multispore isolates (ms) comprise the contents of the whole ascus (Fig. 6.2). Ascospore germina- tion rates were high (7–8 ascospores per ascus germinated). All sporelings revealed the same phenotype and grew equally fast. Fingerprints of genomic DNA derived from five out of eight single spore isolates (1–5) and from sterile cultured vegetative mycelium isolated from the fruiting body (f) revealed the same pattern (Fig. 6.3) with all primers tested. Figs. 6.4–6.6. Thalli of the cross-fertilized (heterothallic) Xanthoria calcicola are irregularly fertile, carrying no, few, to many fruiting bodies (apothecia). Ascospore germination rates were low to medium (0–6 ascospores per ascus germi- nated). The maximum 6 single spore isolates obtained per ascus revealed different phenotypes and growth rates (Fig. 6.5). Fingerprints of genomic DNA varied (arrows point to 2 polymorphic markers in Fig. 6.6). but species with numerous ascomata were either homothallic (e.g. X. parietina, X. elegans), or heterothallic (e.g. X. polycarpa, Physcia aipolia, Ramalina fastigiata) (Honegger et al. 2004; Honegger and Zippler 2007). Phylogenetic analyses combined with studies on mating systems give an insight into evolutionary and speciation processes. The widespread and common Xanthoria parietina was assumed to be the primary species, from which the European X. calcicola is derived (Purvis et al. 1992). Xanthoria flammea, a morpholo- gically interesting South African endemic, was placed in the monotypic genus Xanthomaculina (Ka¨rnefelt1989). Xanthoria calcicola and X. flammea are heterothallic, 98 R. Honegger and S. Scherrer
X. parietina is homothallic (Honegger et al. 2004; Scherrer et al. 2005). Comparative analyses of the noncoding rDNA region (ITS 1 and 2, 5.8 S) and of the hydrophobin gene sequences revealed a relatively close relationship of X. flammea with the X. parietina complex, and the homothallic X. parietina was found to be a derived, not a primary, species (Scherrer and Honegger 2003; Scherrer et al. 2005).
6.2 Dikaryon formation
Dikaryon formation (pairing, but not fusion of nuclei), a peculiarity of ascomycetes and basidiomycetes, remains largely unexplored in lichen-forming taxa. Nonlichenized ascomycetes produce dikaryons either by gametangial fusion (ascogonium with antheridium), spermatization (gamete fusion with gametangium, i.e. fusion of microconidia or macroconidia with the trichogyne of the ascogonium), or fusion of undifferentiated vegetative hyphae and subse- quent pairing of nuclei (somatogamy). In thalli of lichen-forming ascomycetes ascogonia are formed within or slightly underneath the algal layer. In many species trichogynes are seen pro- truding above the thallus surface (Figs. 6.8–6.9), which is often locally covered by mucilaginous material at this particular site (examples in Jahns 1970; Henssen and Jahns 1973; Honegger 1978a, b). The majority of lichen-forming ascomy- cetes produce large numbers of tiny, often bacteria-sized microconidia in flask- shaped conidiomata with narrow ostiole (pycnidia; Figs. 6.10–6.13), whose cavity is filled with a hydrophilic mucilage. In wet weather the swelling mucilage causes masses of microconidia to ooze out of the narrow apical ostiole (Fig. 6.11), whence they are most likely dispersed either by invertebrate vectors, as in many nonlichenized fungi such as Claviceps purpurea, and Epichloe¨ spp., or by rusts (summarized by Naef et al. 2002) or by rain (splash dispersal). In many species pycnidia are well exposed, e.g. in projections along the thallus margin of Cetraria islandica (‘‘Icelandic moss’’) and other Cetraria spp., or at the tip of podetia in Cladonia spp. (reindeer and cup lichens; Figs. 6.7, 6.10). In most foliose species (Parmeliaceae, Physciaceae, Teloschistaceae, etc.), pycnidia are immersed, but the ostiole opens at the surface of the lobes. Spermatization is assumed to be the mode of dikaryon formation in the numerous species of lichen-forming ascomycetes, in which microconidia have been seen adhering to protruding trichogynes (Jahns 1970; Henssen and Jahns 1973; Honegger 1978a, 1984a, b). In Cladonia furcata microconidia adhere, tip first, to the surface of trichogynes (Fig. 6.9), dissolve their own and the trichogyne wall at the contact site and leave a hole in it (Honegger 1984a). Trichogynes are short-living, ephemeral receptive hyphae which die off once that dikaryotization is completed. Sexual reproduction in lichen-forming ascomycetes 99
Figs. 6.7–6.13 Spermatization in Cladonia furcata. Fig. 6.7. Topmost fragment of a fertile thallus. Details of framed areas are shown in Figs. 6.8 and 6.10.Fig.6.8.Group of three ascomatal primordia with numerous trichogynes (tr). Fig. 6.9. Detail of a trichogyne (tr) and adhering microconidia (c). Arrow points to a microconidium (¼ spermatium) that is attached, tip first, to the wall of the trichogyne. Fig. 6.10. Pycnidium (py) at the tip of the branch. Detail of framed area is in Fig. 6.11.Fig.6.11. Arrow points to the pycnidial ostiole. Innumerable microconidia (c) were oozing out of the pycnidial cavity. Fig. 6.12. Longitudinally dissected pycnidium after removal of the hydrophilic mucilage that filled the pycnidial cavity. Conidiophores grow out of the pycnidial wall and release microconidia into the pycnidial cavity. Detail of framed area is in Fig. 6.13. Fig. 6.13. Mature microconidium (c) detached from the conidiophore (cp). Arrows point to conidiophores with developing microconidia. (Figures 6.8 and 6.13 are from Honegger 1984b, with permission of the publisher.)
Migration of conidial nuclei through the trichogyne and pairing with ascogonial nuclei likely occurs, but has never been documented in lichen-forming ascomy- cetes. In many species ascogonia were never found with protruding trichogynes (e.g. in the homothallic Xanthoria parietina; Janex-Favre and Ghaleb 1986; con- firmed with SEM techniques by R. Honegger, unpublished.); others produce large numbers of trichogynes but neither macroconidia nor microconidia, which might serve as gametes (e.g. Peltigera spp.; R. Honegger, unpublished.). In the vast majority of species ascomal primordia and microconidiomata (pycnidia) are differentiated in the lichenized part of the thallus, usually within the algal layer. However, in Rhizocarpon spp. and possibly in other taxa of 100 R. Honegger and S. Scherrer
crustose lichen-forming ascomycetes, ascomal primordia and microconidio- mata develop in the prothallus of thalli adjacent to lichenized areoles (Honegger 1978a).
6.3 Ascomal ontogeny
The stimuli triggering ascomal initiation and differentiation are unex- plored in lichen-forming ascomycetes. As in most nonlichenized taxa the vege- tative hyphae surrounding the dikaryotic, ascogenous hyphae build up a fruiting body with all elements characteristic of the taxon (see Chapter 17; Henssen and Jahns1973 ; Henssen198 1; Parguey-Leduc and Janex-Favre1981 ). When an ascogenous hypha has reached the hymenial layer it differentiates, after crozier formation and thus ultimate distribution of complementary nuclei, an ascus, in which the paired haploid nuclei fuse (i.e. undergo karyogamy) to form a diploid nucleus, the zygote. In the young ascus the zygote undergoes meiosis, resulting in four haploid nuclei, which may subsequently go through one or more mitotic nuclear divisions.
6.4 Ascosporogenesis
From a cell biological point of view, ascospore formation is a very interesting process, during which new cells are formed within the protoplast of an existing one (free cell formation). As in nonlichenized ascomycetes, the postmeiotic ascus of lichen-forming taxa comprises four haploid nuclei, which divide once, two times or even more, resulting in 8, 16, 32, or 64, etc. nuclei per ascus, 8 being found in the majority of species. A double membrane sac, often termed the peripheral membrane cylinder, is differentiated near the periphery of the protoplast. As in nonlichenized ascomycetes (Beckett 1981) different membrane systems were found to contribute to the formation of this membrane sac: the plasma membrane in Peltigera spp. (Fig. 6.14), Baeomyces rufus or Chaenotheca chrysocephala (Honegger 1982b, 1985), or the endoplasmatic reticu- lum in Physcia stellaris, Pleurosticta acetabulum or Rhizocarpon geographicum (Honegger 1982b). Guided by microtubules of the spindle apparatus, the double membrane sac invaginates around nucleate portions of cytoplasm and finally ruptures at the invagination fronts, thus becoming fragmented into small sacs, each of which is a future ascospore. Ascospore delimitation is completed upon closure of each of these small membrane sacs (Fig. 6.15). The inner membrane is the plasma membrane of the ascospore, the outer one the investing membrane, which is in contact with the cytoplasm of the ascus; it breaks down in the mature spore after wall completion. Sexual reproduction in lichen-forming ascomycetes 101
Figs. 6.14–6.16 Ascosporogenesis in Peltigera canina, as seen in transmission electron micrographs of longitudinally sectioned asci that had been chemically fixed with potassium permanganate. Fig. 6.14. Differentiation of a peripheral membrane sac, the ascus vesicle (av), in the premeiotic ascus via blebbing of the plasma membrane (bold arrows) near the ascus tip, where the ascus wall (aw) is thickened. In the postmeiotic ascus the ascus vesicle envelops nucleate portions of cytoplasm, the future ascospores. Figs. 6.15–16. Asci after ascospore delimitation (Fig. 6.15) and ascospore maturation (Fig. 6.16). Asci have an apically thickened wall (aw) and an eversible apical ring (ar) that stains blue with iodine in light microscopy preparations. Asci and paraphyses (p) are embedded in the hydrophilic hymenial gelatine (g) secreted by the paraphyses, which swells dramatically in the wet state, thus gener- ating a high pressure on the flanks of the asci. Empty asci (a) after ascospore ejection are tangentially sectioned. The topmost part of expanding asci reached the hymenial surface (hs) during ascospore release. At maturity the fusiform, multicellular asco- spores (asp) comprise one nucleus (n) per cell and numerous lipid-storing vesicles (l). Magnification bars represent 1 mm.
Large numbers of genera and species of lichen-forming ascomycetes have one-celled ascospores. However, within the freshly delimited ascospore nuclear divisions, followed by cell divisions, may occur according to species- or genus- specific parameters, leading to 2-, 4- or multicelled spores (Fig. 6.16). Under favorable conditions each cell of the ascospore germinates. Multicellularity increases the chance of survival under difficult conditions. During ascospore wall differentiation, at least two layers are deposited. The innermost corresponds in its 102 R. Honegger and S. Scherrer
structure and composition to the hyphal wall and will form a continuum with the wall of the germ tube. Outer wall layers may be locally thickened and hyaline or pigmented, e.g. gray to black due to inclusion of melanin granules (e.g. Rhizocarpon or Physcia spp.; Honegger 1978b, 1980). Several genera of lichen- forming ascomycetes from extreme climates (arctic, antarctic and high alpine ecosystems) such as rock tripes (Lasallia and Umbilicaria spp.) or map lichens (Rhizocarpon spp.) have muriform (multicellular) ascospores with strongly mela- nized walls. Melanin protects from UV radiation and possibly prevents loss of soluble compounds (Butler and Day 1998). Multicellularity enhances survival rates, one or few cells of the spore staying viable while others die under harsh conditions. An exceptional situation is found in Pertusaria spp. with either one or two large, multinucleate ascospores per ascus, which presumably comprise all, as in the unisporate P. bryontha, or half of the postmeiotic nuclei as in the bisporate P. pertusa.
6.5 Ascus structure and function
Most lichen-forming ascomycetes develop their asci in a hymenium, which is filled with a hydrophilic mucilage (Figs. 6.15, 6.16). This mucilage is produced and released by the paraphyses (Figs. 6.15, 6.16), vegetative, haploid hyphae with characteristic growth patterns. Tip cells of the paraphyses may also secrete secondary compounds, which crystallize at the hymenial surface and give the apothecial disk its characteristic color. Paraphyses of some species secrete the same secondary compounds as cortical vegetative hyphae in the thallus (e.g. yellow to orange anthraquinones in numerous Teloschistaceae) or different ones (e.g. the blood red naphtaquinone haemoventosin in the grayish Ophioparma ventosa, or the bright red anthraquinone bellidiflorin in the greenish gray Cladonia bellidiflora; Hunek and Yoshimura 1996). During hydration the ascomal mucilage swells substantially and generates a high pressure on the flanks of the asci. In the course of ascospore maturation the ascus wall under- goes a series of differentiation processes, which culminate in ascospore release. These differentiation processes include deposition of wall layers and, in species which eject ascospores, differentiation of a characteristic apical apparatus (Fig. 6.16; see Fig. 4.36 in Chapter 4). The apical apparatus of immature asci withstands the high pressure generated by swelling of the hydrophilic mucilage within the hymenial layer and stays closed. When ascospores are nearly mature, defined zones of the apical apparatus change their structural integrity and finally rupture, a prerequisite for successful ascospore discharge. In species with rostrate ascus dehiscence (see4.36B Fig.–D in Chapter4 ), the rupture of outer wall layers at the apex facilitates the tube-like expansion of inner ones, Sexual reproduction in lichen-forming ascomycetes 103 which reach the hymenial surface or expand even above. This rostrate dehis- cence is typically found in Lecanoraceae, Cladoniaceae, Physciaceae, etc. (Honegger 1978b). Many Peltigeraceae carry an eversible amyloid ring at the tip of their expansible inner wall layer (Figs. 6.16, 4.36A; Honegger 1978b). Mazaediate ascomata are not filled with mucilage and their asci do not eject the spores, as the ascospore wall simply disintegrates during ascospore matura- tion (Honegger 1985), leading to a powdery ascospore mass on the surface of the fruiting body. Ascus structure and function and features of ascomal ontogeny are widely used as taxonomic characters. Phylogenies based on molecular markers are largely in agreement with these sets of data, but partly show that structurally and functionally similar ascus types and ascomata have evolved independently in different groups. An example is the genus Sphaerophorus, formerly included in Caliciales on the basis of its mazaediate ascomata and deliquescent asci (Henssen and Jahns 1973), but now recognized as a family Sphaerophoraceae within the Lecanorales (Wedin and Do¨ ring 1999). The same applies for Caliciaceae, now included in Lecanorales, the Caliciales having been omitted (Eriksson et al. 2006b). 7 Biochemistry and secondary metabolites
j. a. elix and e. stocker-wçrgçtter
7.1 Intracellular and extracellular products
There are two main groups of lichen compounds: primary metabolites (intracellular) and secondary metabolites (extracellular). Common intracellu- lar products occurring in lichens include proteins, amino acids, polyols, car- otenoids, polysaccharides, and vitamins, which are bound in the cell walls and the protoplasts, are often water-soluble, and can be extracted with boiling water (Fahselt 1994b). Some of these products are synthesized by the fungus and some by the alga. Since the lichen thallus is a composite structure, it is not alwayspossibletodecidewhereaparticularcompoundisbiosynthesized. Most of the intracellular products isolated from lichens are nonspecific, and also occur in free-living fungi, algae and in higher green plants (Hale 1983).The majority of organic compounds found in lichens are secondary metabolites of the fungal component, which are deposited on the surface of the hyphae rather than within the cells. These products are usually insoluble in water and can only be extracted with organic solvents. Carbon for the lichen is furnished primarily by the photosynthetic activity of the algal partner. Mosbach (1969) summarized the overall carbon metabolic sequence as invol- ving photosynthesis in the photobiont followed by transport of the carbo- hydrate to the fungus, metabolism of the carbohydrate and subsequent biosynthesis of lichen secondary metabolites. The type of carbohydrate released by the alga and supplied to the fungus is determined by the photo- biont, while in lichens containing cyanobacteria, the carbohydrate released and transferred to the fungus is glucose. In lichens containing green algae, the carbohydrate released and transferred to the fungus is a polyol: ribitol, ery- thritol, or sorbitol (Section 10.2.1).
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. Biochemistry and secondary metabolites 105
7.2 The fungal origin of the secondary metabolites
All of the secondary substances which are so characteristic of lichens are of fungal origin. Consequently it seems rather surprising that with more than 700 secondary metabolites known from lichens (Huneck 1999; Dembitsky and Tolsikov 2005), most are unique to these organisms and only a small minority (c. 50–60) occur in other fungi or higher plants. As an example, the anthraquinone parietin, the orange pigment common in most Teloschistales, occurs in species of the non- lichenized fungal genera Achaetomium, Alternaria, Aspergillus, Dermocybe, Penicillium, as well as in the vascular plants Rheum, Rumex,andVentilago. Similarly, the common para-depside lecanoric acid also occurs in the fungal genus Pyricularia, while the typical higher plant sterol, brassicasterol, has also been detected in the lichens.
7.3 Biosynthetic pathways to lichen secondary metabolites
Direct evidence from biosynthetic investigations on intact lichens using labeled compounds is meager, but hypothetical pathways are often pro- posed on the basis of what is known for the biosynthesis of analogous fungal products (Turner and Aldridge 1983). In addition, further circumstantial evi- dence may be forthcoming from observed joint occurrence of compounds and laboratory interconversions and biomimetic syntheses (C. Culberson and Elix 1989). In the past nearly all the chemical data came from studies of natural lichens because cultures of lichen fungi grow very slowly and failed to show all products characteristic of mature thalli in nature. However, recent advances in the controlled growth of recombined species promise to open new areas of research whereby the biosynthetic sequence to various lichen acids can possibly be confirmed (Section 7.5; Hamada et al. 2001). Most of the secondary metabo- lites present in lichens are derived from the acetyl-polymalonyl pathway, but some come from the shikimic acid and mevalonic acid pathways (C. Culberson) and Elix 1989; Huneck 2001). An overview of the probable biosynthetic path- ways to the major classes of lichen products is illustrated in Fig. 7.1. One of the more interesting developments in recent years is the recognition of the key role played by para-depsides as potential precursors (or biosynthetic intermediates) to meta-depsides, depsones, diphenyl ethers, depsidones and dibenzofurans (C. Culberson and Elix 1989). Very recent experimental evidence obtained with cultured lichens is consistent with these suggestions.
7.4 Major categories of lichen products
The first classification of lichen substances based on known structures and biosynthetic pathways was constructed by Asahina and Shibata (1954). This 106 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Fig. 7.1 Probable pathways leading to the major groups of lichen products.
system is modified from time to time, as more information has become avail- able, most recently by C. Culberson and Elix (1989). The principal classes of lichen secondary metabolites are listed in Table 7.1 according to their probable biosynthetic origin, with the approximate number of compounds of known structure indicated (in brackets) and the structure of typical representative compounds (Figs.7.2 –7.5 ). Of the acetyl-polymalonyl derived compounds, aromatic products are especially well represented (Figs. 7.2, 7.3), the most characteristic being formed by the bonding of two or three orcinol or b-orcinol-type phenolic units through ester, ether and carbon–carbon linkages (Fig. 7.3). The large majority of depsides, depsidones, dibenzofurans, usnic acids and depsones all appear to be produced by such mechanisms and all are peculiar to lichens. Other aromatic compounds of acetate-polymalonate origin, such as the chromones, xanthones, and anthraquinones, are probably formed by internal cyclization of a single, folded polyketide chain (Table 7.1) and are often iden- tical or analogous to products of non-lichen-forming fungi or higher plants. In addition to the compounds of known chemical structure, many of unknown structure are given common names and assigned to compound classes, because they are frequently encountered and easily recognized by microchemical methods. Biochemistry and secondary metabolites 107
Table 7.1. Major classes of secondary metabolites in lichens
1. Acetyl-polymalonyl pathway 1.1 Secondary aliphatic acids, esters and related derivatives (45) 1.2 Polyketide derived aromatic compounds 1.2.1 Mononuclear phenolic compounds (19) 1.2.2 Di- and tri-aryl derivatives of simple phenolic units 1.2.2a Depsides, tridepsides and benzyl esters (185) 1.2.2b Depsidones and diphenyl ethers (112) 1.2.2c Depsones (6) 1.2.2d Dibenzofurans, usnic acids and derivatives (23) 1.2.3 Anthraquinones and biogenetically related xanthones (56) 1.2.4 Chromones (13) 1.2.5 Naphthaquinones (4) 1.2.6 Xanthones (44) 2. Mevalonic acid pathway 2.1 Di-, sester- and triterpenes (70) 2.2 Steroids (41) 3. Shikimic acid pathway 3.1 Terphenylquinones (2) 3.2 Pulvinic acid derivatives (12)
7.5 Molecular studies on polyketides and secondary metabolites of lichens (polyketides of lichen-forming fungi)
7.5.1 Origin and distribution of polyketides Polyketides are a class of naturally occurring metabolites found in bacteria (prokaryotes), fungi (lichen-forming fungi; Fig. 7.6), algae, and higher plants, as well as in the animal kingdom (e.g. in dinoflagellates, insects, mollusks, and sponges). Polyketides are usually categorized on the basis of their chemical structures. An immensely rich diversity of polyketide structural moieties have been detected and structurally elucidated, and more await discovery.
7.5.2 Biosynthesis and assembly of polyketides Polyketides are biosynthesized by sequential reactions catalyzed by an array of polyketide synthase (PKS) enzymes. PKSs are large multienzyme protein complexes that contain a typical core of coordinated active sites. The biosynthesis of polyketides occurs stepwise from 2-, 3- and 4-carbon building 108 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Fig. 7.2 Structures of typical acetyl–polymalonyl lichen products derived from a single polyketide chain.
blocks, such as acetyl-CoA, propionyl-CoA, butyryl-CoA and their activated deri- vatives malonyl, methylmalonyl, and ethylmalonyl-CoA (Fig. 7.7). The major polyketide chain building step is a decarboxylative condensation (closely related to the chain elongation step in fatty acid biosynthesis). By chemical and biochemical comparisons, a mechanistic relationship between polyketide and fatty acid biosynthesis has been recognized, whereby the carbon backbones of the respective molecules are assembled by successive condensation of acyl Biochemistry and secondary metabolites 109
Fig. 7.3 Structures of typical acetyl–polymalonyl lichen products derived from two or more polyketide chains. units. Polyketide synthases and fatty acid synthases (FASs) are multifunctional enzymes with a similar ancestral ketoacylsynthase domain (KS), acyltransferase (AT), ketoreductase (KR), dehydratase (DH), enoylreductase (ER), and acyl carrier protein (known as a phosphopantetheine attachment site or PP domain). The KS, AT and PP domains are essential for both FASs and PKSs. Although the KR, DH, and ER are found in all FASs, some or all are absent in PKSs. Ketoreductase, 110 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Fig. 7.4 Structures of typical lichen products derived from the shikimic acid pathway.
DH and ER domains catalyze the stepwise reduction of a keto group to a hydro- xyl group, dehydration of the hydroxyl to an enoyl group, and, finally, the reduction of the enoyl to an alkanoyl group. In the case of fatty acid biosynth- esis, each successive chain elongation step is followed by a fixed sequence of ketoreduction, dehydration and enoylreduction; whereas the individual chain elongation intermediates of polyketide biosynthesis undergo all, some, or none of the functional group modifications. This results in a remarkable diversity of Biochemistry and secondary metabolites 111
Fig. 7.5 Structures of typical lichen products derived from the mevalonic acid pathway.
optional domains
KS AT DH MT ER KR AC CYC
K eto Synthase Acyl Carrier Acyl Transferase K eto Reductase Claisen-type DeHydr atase Eno yl Reductase cyclase
Methyl Transferase Domains with reducing activities
Fig. 7.6 Model of fungal type I PKSs. 112 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Fig. 7.7 Biosynthesis of depsidones and usnic acids (Chooi et al. 2006, IMC8).
structural motifs and levels of complexity of polyketide molecules. Polyketide synthases that lack some or all of these domains produce reduced (e.g. lovastatin formed by Aspergillus terreus), partially reduced, or fully oxidized polyketides. Both types of polyketides are found in lichen-forming fungi, e.g. reduced polyketides, such as bourgeanic acid and anthrones (reduced anthraquinones), and fully oxidized polyketides, such as depsides, depsidones, b-orcinol Biochemistry and secondary metabolites 113 depsidones, and dibenzofurans (Section 7.4). The formation of oxidized polyke- tides (most of the well-known and common lichen polyketides) is controlled by nonreducing PKS genes (Schmitt et al. 2005a).
7.5.3 Types of PKSs, biosynthesis of lichen polyketides, and PKS genes from lichens Recent research has shown that three architecturally different polyke- tide synthases (PKSs) occur in the prokariotic and eukaryotic organismal world (Bedford et al. 1995; Cox et al. 1997; Hopwood 1997; Cane et al. 1998; Bingle et al. 1999; Nicholson et al. 2001). Types I and II are present in bacteria and fungi, have multifunctional enzymes or aggregates of monofunctional enzymes, which operate upon substrates bound by thioester linkages to an acyl carrier protein. Type III PKSs, found in higher plants, lack the ACP moiety, and instead use coenzyme A esters. Type I systems consist of large multifunctional proteins, which can be either noniterative (e.g. modular systems responsible for biosynthesis of macrolides, large ring compounds such as erythromycin, rifamycin, etc.) or iterative (Cane et al. 1998). The iterative type I PKSs are single protein complexes (single mod- ules), that contain all the necessary domains and use their active sites repeatedly
(iteratively) to produce a particular polyketide. They add a C2 molecule (e.g. a CoA ester) to the growing chain with each condensation and cycle repeat. The products of an iterative and noniterative PKS can be joined; and, in this case, result in the formation of a branched PK. The diversity of PKSs results from the use of the three optional PKS reducing domains as described above (Kroken et al. 2003). Iterative type I polyketide synthases, analogous to vertebrate FASs, are typical for the biosynthesis of fungal polyketides, e.g. 6-methylsalicylic acid and afla- toxins. The former has been identified as a precursor of aflatoxins, as well as norsolorinic acid, and other anthraquinones. Anthraquinones are common, polyketide-derived pigments in lichens, and they also occur in nonsymbiotic fungi and higher plants (e.g. Rumex spp.). Another common fungal metabolite is orsellinic acid. In Penicillium griseum, penicillic acid is formed by a gross struc- tural modification of orsellinic acid. Orsellinic acid is also a common precursor of many lichen substances, including depsides and depsidones. To date, PKS genes have been found in clusters (genes adjacent along one stretch of a chromosome). Fungal secondary metabolites are encoded by clusters of sequentially arranged genes (Keller and Hohn 1997). Fungal PKS genes encode multifunctional proteins (fungal type I PKSs) with only one single, reiteratively used ketoacyl synthesis domain, which sequentially condenses C2 units. The gene fragment encoding the ketoacyl domains are highly conserved and can be easily targeted with PKS primers. 114 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Grube and Blaha (2003) and Schmitt et al.(2005a) have utilized a phylogenetic approach to elucidate the relationships between fungal genes by using amino acid sequences of KS (ketoacyl domains) of fungal PKSs, which putatively pro- duce nonreduced (oxidized) polyketides (lichen substances). Such molecular analyses can be useful in reconstructing the evolutionary history of PKS genes in general, but more particularly, in identifying subgroups of type I PKS genes. In particular, they hope to elucidate the evolution of metabolic diversity within selected lichen orders or families that have been chemically characterized. The location and identification of putative PKS genes and FASs are sometimes complicated by the fact that both genes have a strong sequence similarity, reflecting the roles of the enzymes coded by the genes. Other genes, e.g. those coding for cyclases, which catalyze the formation of aromatic polyketides (typi- cal for lichens) have no counterparts among the genes coding for FAS (forming molecules arranged in long chains).
7.5.4 Heterologous expression of a lichen PKS in filamentous fungi Although filamentous fungi produce an immense variety of polyketides (Simpson 1995), only a few PKS genes have been isolated (Sinnemann et al. 2000). The first PKS gene from a lichen has recently been isolated and sequenced by O. Andre´sson and S. P. Davidsson (pers. comm., EUKETIDES Meeting, 2006). This PKS gene was obtained from Solorina crocea, then cloned and expressed in several filamentous fungi (Aspergillus nidulans, A. niger, A. oryzae, and Fusarium venenatum) by standard cloning and recombination technique. A 16 kb plasmid with a marker mediating hygromycin resistance was constructed, and together with a strong fungal promoter, the transcription of the lichen PKS gene was achieved. Further genetic transformation of A. niger with this plasmid construct yielded transformants that were able to produce a pigment of yet unknown chemical structure (from Gagunashvili and Andre´sson, 2006). Such experiments hope to produce polyketides and polyketide-type pigments of considerable potential for drug discovery and novel biological activity.
7.5.5 Fatty acids and polyketides in lichens and cultured lichen fungi (mycobionts) In many of the earlier investigations (Yamamoto 1990; Ahmadjian 1993; Kinoshita 1993), the majority of cultured mycobionts did not produce polyke- tides, i.e. typical lichen metabolites. Interestingly, when polyketides were pro- duced in culture, alternative substances were often formed rather than those present in the original lichen or voucher specimens. The observed results were often difficult to interpret, and factors which favored the production of lichen substances remained unrecognized for several decades. More recently, it was shown that lichen mycobionts, which do not produce polyketides, may Biochemistry and secondary metabolites 115 biosynthesize fatty acids instead. Molina et al.(2003) found that axenic cultures of Physconia distorta grown on nutrient-rich media produced mainly fatty acids (oleic, linoleic, and stearic acids) and their triglyceride derivatives, substances which were deposited on the surface of the mycelia as fat droplets. These experiments showed that FAS (fatty acid synthase) was switched on and acti- vated, whereas PKS was obviously inhibited. In another study (Adler et al. 2004), an aposymbiotically grown mycobiont, cultured under stable culture conditions, did not produce the typical medullary polyketide gyrophoric acid, but instead generated hydrocarbons, monoacylgly- cerides, and triacylglycerides. Such metabolic switching has also been observed in filamentous fungi, such as Aspergillus nidulans (Archer et al. 1999). Heterologous expression and cloning yielded both PKS and FAS genes. If PKS and FAS genes do not form separate gene clusters, the search for the location of putative PKS genes can be very difficult and needs advanced molecular genetic methods. In this case, the research challenge is to locate the genes that control the biosynth- esis, modification, and in some cases secretion and resistance of polyketides. Then one could clone the PKS genes so that they can be moved to and be expressed in well-defined cell factories, like Aspergillus niger or A. oryzae. A recent approach (Brunauer and Stocker-Wo¨rgo¨tter2005; Stocker-Wo¨rgo¨tter 2005, 2008; Brunauer et al. 2006) was undertaken to sequence a cDNA of the PKS for anthraquinone production. The lichen and especially the cultured mycobiont of Xanthoria elegans were found to be excellent model organisms, as the aposymbiotically grown mycobiont readily produced anthraquinones, like parietin, teloschistin, etc. This meant that the mRNA was actually transcribed. From the axenically grown mycobiont, clean RNA was isolated, and then used for synthesis of cDNA by utilizing the SMART RACE cDNA synthesis technique. The SMARTTM technology provided an efficient method for producing a cDNA pool, enriched in full-length cDNA and incorporating primer binding sites, at the 50- and the 30-ends of the cDNA, following 50-and30-RACE (Rapid Amplification of cDNA Ends)-PCRs (Zhu et al. 2001). To obtain specific amplification of the PKS cDNA, gene specific primers (GSP) were designed based on the known sequence of the KS domain of the enzyme and used together with the oligos for the incorpo- rated primer binding sites at the 50- and 30-ends of the cDNA. The resulting amplicons of the 50- and 30-RACE-PCRs were cloned into a T-vector and sequenced. The cDNA sequence was then analyzed using the ORF prediction program implemented in VestorNTI. The resulting amino acid sequence was then subjected to a Blast search against the NCBI database. The Blast search revealed high homology to other known PKS enzymes, especially to the wA gene product (Accession: Q03149; Fujii et al. 2001)ofAspergillus nidulans, which pro- duces a polyketide structurally homologous to the polyketides of X. elegans. 116 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Several catalytic domains on the enzyme could be identified and, finally, a gene bank for PKS of the Xanthoria elegans fungus was established. Chooi et al.(2006) searched for a PKS gene responsible for production of b- orsellinic acid and methyl-phloroacetophenone, as precursors of typical lichen polyketides (depsidones and usnic acids) in the Australian lichen Chondropsis semiviridis. They found that both genes, either controlling depsidone and/or usnic acid production, were probably coregulated and were part of a common, larger gene cluster. In this case, the polyketide gene was identified by hetero- logous expression in a surrogate host, e.g. Aspergillus nidulans. For this procedure, the PKS gene clones of Chondropsis semiviridis were transformed and a strong promoter in the chosen transformation host was found. Today heterologous expression of PKS genes and successive production of larger quantities of biologically active polyketides (e.g. actinomycetous com- pounds and fungal metabolites) are becoming common strategies to obtain and design pharmaceutically useful molecules. Probably in the near future similar molecules will also be found in lichens (Boustie and Grube 2005).
7.6 Detection and identification of secondary lichen substances
7.6.1 History The application of chemical discriminators to lichen taxonomy began inadvertently when thalline color was accepted as a generic or specific charac- ter. Hence the gray-green genus Physcia (containing the colorless substance atranorin in the cortex) was segregated from the superficially similar but yellow-orange genus Xanthoria (containing cortical parietin, an orange pigment). Similarly Parmeliopsis ambigua (with a yellow thallus due to the presence of usnic acid) was separated from the morphologically similar P. hyperopta (gray with atranorin). Nevertheless, most lichen substances are colorless and can be detected only by indirect means. The first chemical tests conducted on lichen thalli for taxonomic purposes were carried out by Nylander in the 1860s (Nylander 1866; Vitikainen 2001). He detected the presence of various colorless lichen substances by spotting chemical reagents directly on the lichen thallus (spot tests) to produce characteristic color changes. He used solutions of iodine, potassium hydroxide (K), and calcium hypochlorite (C). Further test reagents followed – KC (K solution followed by C) and CK (with reverse addition) – but the origin of these characteristic color reactions remained unknown. The first exten- sive chemical investigations on lichens were conducted by Hesse and Zopf, culminating in Zopf’s (1907) publication of Die Flechtenstoffe,inwhichdescriptions of over 150 lichen compounds appeared. The ultimate structural elucidation of Biochemistry and secondary metabolites 117
Table 7.2. Reagents for thalline spot tests
K ¼ 10% aqueous KOH solution a. Turns yellow then red with most o-hydroxyl aromatic aldehydes. b. Turns bright red to deep purple with anthraquinone pigments.
C ¼ saturated aqueous Ca(OCl)2 or common bleach (NaOCl) solution a. Turns red with m-dihydroxy phenols, except for those substituted between the hydroxy
groups with a -CHO or -CO2H. b. Turns green with dihydroxy dibenzofurans.
KC ¼ 10% aqueous KOH solution followed by saturated aqueous Ca(OCl)2 or common bleach (NaOCl) solution a. Turns yellow with usnic acid. b. Turns blue with dihydroxy dibenzofurans. c. Turns red with C- depsides and depsidones which undergo rapid hydrolysis to yield a m-dihydroxy phenolic moiety.
PD ¼ 5% alcoholic p-phenylenediamine solution a. Turns yellow, orange or red with aromatic aldehydes.
many common lichen metabolites was due to the subsequent meticulous work of Asahina and coworkers in Japan during the 1930s (Asahina and Shibata 1954). This laid the foundation for further research on these compounds in recent times. Methods were recently summarized by Huneck and Yoshimura (1996).
7.6.2 Localization of secondary products Thalline spot tests and the distribution of pigments provided the first evidence that the lichen substances were not distributed evenly throughout the thallus. In some species the striking red or orange anthraquinone derivatives and the yellow pigment usnic acid were obviously restricted to the upper cortex. Similarly spot tests demonstrated that many of the colorless depsides and depsidones were restricted to the medullary layer. The common spot test reagents (summarized in Table 7.2) not only indicate where particular com- pounds are located in sectioned thalli, but may also give a clue to the chemical nature of the substance. In recent times scanning electron microscopy and laser microprobe mass spectrometry have been utilized to identify particular crystals present on or in the thallus. For instance, scanning electron microscopy (SEM) of the cortex of Lecanora cerebellina Poelt showed crystals with two morphologies. This species is known to contain the chlorinated xanthone, vinetorin. Luminescent crystals showing a signal for chlorine by energy-dispersive X-ray spectrometry (EDX) 118 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
were identified as vinetorin, while nonluminescent crystals of different mor- phology gave a strong signal for calcium but little chlorine, and were tentatively identified as calcium oxalate. X-ray diffraction analysis of the crystals picked from the surface of Pyxine caesiopruinosa (Nyl.) Imsh. confirmed that the abundant bipyramidal crystals were calcium oxalate dihydrate and not the secondary product lichexanthone, known to occur in this species. Crystals have also been identified by mass spectrometry using an instrument that combines a light microscope and a microprobe mass spectrometer with a lateral resolution of about 1 mm. Because ionization is laser induced, this method is also potentially applicable to thermally labile compounds. Laser microprobe mass spectrometry (LMMS) of solid inclusions in a cross section of the thallus of Lauerera benguelensis (Mu¨ ll. Arg.) Zahlbr. confirmed the presence of lichexanthone by the prominent (M þ Hþ) peak at m/z 287 in the positive ion spectrum. LMMS has also been coupled with fluorescence microscopy and transmission electron microscopy (TEM) to locate compounds in semithin sec- tions of several species. The point analyzed on a laser microprobe is then examined in detail by TEM. For example, lichexanthone and russulone, a new tetracyclic anthraquinone, were located in different zones of the fruiting body of Lecidea russula Ach. (C. Culberson and Elix 1989).
7.6.3 Microchemical methods Although the structural elucidation of lichen compounds results from a combination of classical chemical methods and modern spectroscopic techni- ques, the extensive data on the natural occurrence of these compounds are primarily based on microchemical methods of analysis. Extensive surveys based on extracts from herbarium specimens began in the 1930s when Asahina developed a simple microcrystallization technique to identify particu- lar compounds, most of which were of known chemical structure and were located in particular histological regions of the thallus.
7.6.4 Microcrystallization Asahina’s microcrystallization technique allowed definitive recogni- tion of individual lichen acids on a routine basis (Orange et al. 2001). This simple technique required no special equipment, and with experience generally yielded accurate analyses of major products. It involved extraction of a lichen fragment with acetone, evaporation of the solvent and recrystallization of the remaining residue from a suitable solvent – all conducted on a microscope slide (Asahina and Shibata 1954). A particular lichen substance crystallized in a distinctive shape and color and was identified by comparison with photographs of authentic materials (Hale 1974). Nevertheless, it soon became evident that Biochemistry and secondary metabolites 119 this method could not detect minor components and was inadequate for the study of mixtures. This method is now superseded by more accurate and sensi- tive chromatographic methods. Even so, using Asahina’s method, botanists discovered extensive correlations between chemistry, morphology, and the geographic distribution of lichens.
7.6.5 Paper and thin layer chromatography In the period 1952–56 the Swedish chemist Wachtmeister introduced paper chromatography to identify lichen acids and their hydrolysis products (Wachtmeister 1956; Elix 1999). This method established that the chemistry of many species was more complex than was indicated by microcrystallization techniques. Experimental problems, poor spot resolution, low sensitivities, and long analysis times were subsequently overcome by the development of thin layer chromatography (TLC), which is now the most widely used method for identifying lichen products. This technique improved vastly the speed and certainty of recognition of lichen substances by means which are simple to use and relatively inexpensive.
7.6.6 Standardized TLC methodology A standardized method developed by C. Culberson and coworkers remains in general use. It uses commercially available silica gel TLC plates and employs three solvent systems (designated A, B and C) and two internal controls
(atranorin and norstictic acid), to which all Rf data are compared (C. Culberson 1972; C. Culberson and Amman 1979; C. Culberson et al. 1981). An acetone extract of the lichen is spotted on the plate and subsequently eluted in each solvent system. For each solvent system, a spot is assigned to an Rf class deter- mined by its position relative to the controls. Data on punched cards or compu- ter are then sorted to find all the compounds with the same Rf classes. Of these possibilities, those with similar spot characteristics (color, fluorescence, etc.) are compared chromatographically with the unknown. Additional solvent sys- tems and visualizing agents are available for compounds that do not separate well in the initial analysis, and two dimensional TLC exhibits considerably improved Rf discrimination of structurally similar compounds and enables the identification of minor constituents present in complex mixtures (C. Culberson and Johnson 1976).
The solvent systems Solvent A Toluene-dioxane-acetic acid (180:45:5) is reputed to owe its distinctive characteristics to the ability of dioxane to associate with phenolic hydroxy groups. 120 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Solvent B Hexane-methyl tert.-butyl ether-formic acid (140:72:18) gives good separation of compounds that differ only slightly due to the length of side chains or the number of C-methyl substituents. Solvent C Toluene-acetic acid (170:30) is an excellent general solvent for a wide variety of different compounds. Solvent E Cyclohexane-ethyl acetate (75:25) is recommended for less
acidic compounds that have high Rf values in solvents A, B, and C (e.g. many pigments, esters, triterpenes; Elix et al. 1988; Elix and Ernst-Russell 1993). Solvent G Toluene-ethyl acetate-formic acid (139:83:8) is particularly
useful in separating compounds with relatively low Rf values in solvents A, B and C (e.g. b-orcinol depsidones, secalonic acids).
Visualization of spots One of the most useful features of TLC is the broad range of spot
characteristics that can be used in addition to Rf data. Before spraying, the dried plates are examined in daylight for pigments and for fluorescence or quenching under short and long wavelength ultraviolet (UV) light. Subsequently the plates are sprayed with 10% sulfuric acid and then heated at 110 8C in an oven or hotplate for 10 minutes to develop the spots. Sulfuric acid charring detects the broadest range of compound types, including virtually all terpenes and phenolic derivatives (Fig. 7.8). After the plates are charred, the compounds give a range of characteristic visible colors and some even have a characteristic fluorescence (White and James 1985). More recently a standardized TLC analysis procedure was developed to take advantage of computer technology (Elix et al. 1988; Mietzsch et al. 1993). This method utilizes six solvent systems and eight control compounds. Measured
relative Rf values are used to sort within a computerized database. The programs also list biosynthetically related compounds as an aid to the identification of minor satellite substances.
High performance thin layer chromatography (HPTLC) This modification of the standard TLC method utilizes TLC plates com- prising a thinner layer of smaller grained silica particles (average 5–6 mm compared with 10–12 mm for ordinary TLC plates). It is reported to be a more sensitive method, requiring shorter run times and less solvent but is much more sensitive to humidity than the standard method (Arup et al. 1993). Biochemistry and secondary metabolites 121
Fig. 7.8 Tracing of a TLC plate of extracts from Xanthoparmelia species in TA
(toluene, 200: acetic acid, 30; solvent C). Compounds listed in decreasing Rf.1, X. barbatica (usnic acid, barbatic acid, 4-O-demethylbarbatic acid); 2, X. notata (usnic acid, 4-O-methylhypoprotocetraric acid, notatic acid); 3, X. scabrosa (usnic acid, loxodin, norlobaridone); 4, X. metastrigosa (usnic acid, hypostictic acid, hypoprotocetraric acid, hypoconstictic acid); 5, X. terrestris (usnic acid, norstictic acid, salazinic acid); 6, X. tegeta (usnic acid, stictic acid, constictic acid); 7, X. hypoprotocetarica (usnic acid, hypoprotocetraric acid); 8, X. pertinax (usnic acid, succinprotocetaric acid, fumarprotocetraric acid); standard mixture (atranorin, norstictic acid). 122 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
7.6.7 High performance liquid chromatography (HPLC) Isocratic and gradient elution All of the aromatic lichen products are ideally suited for analysis by high performance liquid chromatography (Elix et al. 2003). Since the advent of bonded reverse phase columns, this technique provides a powerful complement to the established TLC methods. Samples are dissolved in methanol and injected into the appropriate partition column, through which an appropriate solvent or sequence of solvents is passed under high pressure. The substances separate and are detected using a UV detector. The retention time (Rt, or time of passage) and peak intensity are recorded by a chart recorder (Fig. 7.9). HPLC is also used to measure either absolute or relative concentrations of lichen compounds, because the peak intensity (area under the curve) is proportional to the concentration. Earlier applications utilized isocratic elution (an eluant of constant composition) to achieve excellent separations of a variety of depsides, depsidones, and diben- zofuran derivatives (C. Culberson et al. 1979; Lumbsch and Elix 1985). However, gradient elution methods (using a sequence of solvent mixtures) are more effi- cient in analyzing crude lichen extracts, which often contain compounds of wide- ranging hydrophobicities. For example, a 30-minute linear gradient from 0.5% acetic acid in water to 100% methanol was used to separate six known (constictic, stictic, norstictic, psoromic, gyrophoric, and rhizocarpic acids) and seven uniden- tified components in the Rhizocarpon superficiale group (Geyer et al. 1984). More recently Archer and Elix (1993) used a 35-minute gradient from 30% aqueous methanol containing 0.7% ortho-phosphoric acid to 100% methanol to distinguish ten components in an undescribed Pertusaria species (Fig. 7.9). Most workers using HPLC to detect lichen compounds combine this techni- que with TLC and/or mass spectrometry to verify the identification of the peaks. This has often proved to be a bonus, because the unique chemistry of reverse phase separations and the high sensitivity of UV detectors led to the discovery of many compounds new to science. Nevertheless, verification of the identity and purity of peaks remains a problem in screening large numbers of specimens.
Retention indices New standardized methods that use retention indices relative to two internal standards (one of low and one of high retention) rather than retention times, avoid problems caused by column age and minor variations in solvent composition (Huovinen et al. 1985; Feige et al. 1993). Retention indices also provide structural information; for example, C. Culberson et al.(1984) showed that there was a linear relationship between the retention index and the num- ber of side-chain carbon atoms in homologous series of naturally occurring orcinol depsides and their hydrolysis products. Thus, this method can provide Biochemistry and secondary metabolites 123
Fig. 7.9 Trace of HPLC of methanol extract of Pertusaria. sp. (horizontal scale in minutes). A, benzoic acid (internal standard); B, 4,5-dichloronorlichexanthose; C, arthothelin; D, asemone; E, thiophanic acid; F, 3-O-methylasemone; G, 6-O-methyl- asemone; H, superlatolic acid (internal standard). the first clue to the identity of new products (particularly satellite compounds) in lichens. Because of the expense and technical complexity of HPLC, TLC will probably continue to be more widely used in routine identifications. Even so, HPLC is the method of choice for detecting trace satellite compounds, analyzing very small samples, quantifying the lichen products present, and for providing structural information from retention characteristics.
7.6.8 Chemical methods As with many areas of natural product chemistry, new impetus in the chemistry of lichen substances is provided by the more rapid and improved methods for detecting, isolating, and purifying these compounds and in deter- mining their structure. The techniques of preparative TLC, radial chromatogra- phy and preparative HPLC provide rapid and efficient methods for the purification of lichen substances, and developments in mass spectrometry, proton and carbon-13 NMR (nuclear magnetic resonance) spectroscopy and 124 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
X-ray crystal analysis greatly aide structural studies. For details of these methodologies, modern texts on organic structure determination should be consulted. The more classical chemical procedures of degradation and total synthesis also developed apace with the use of newer reagents and synthetic methods. For instance, the use of the condensing reagents trifluoroacetic anhydride and dicyclohexylcarbodiimide make the preparation of lichen depsides a relatively straightforward procedure, so that total synthesis is now a common means of structural confirmation.
7.6.9 Gas chromatography and lichen mass spectrometry As the typical lichen depsides and depsidones contain thermally labile ester linkages, techniques that require volatilization may give decomposition products that complicate the analysis. However, xanthones, anthraquinones, dibenzofurans, terpenes, and pulvinic acid derivatives lack such linkages and have been successfully studied by gas chromatography (GC), gas chromatogra- phy coupled with mass spectrometry (GCMS) and lichen mass spectrometry (LMS). Santesson (1969) was the first to study xanthone pigments using LMS, by introducing small lichen samples (some less than 50 ng) into the direct inlet system of a mass spectrometer. The xanthone sublimes as the temperature is raised (100–150 8C) under very low pressure and the mass spectrum is recorded. Xanthones generally give prominent molecular ions and the spectra of mixtures can often be seen as additive of the individual components (Fig. 7.10; Elix 1999). Aptroot (1987) was able to resolve and identify the main terpenoid components of many lichens of the Pyxinaceae by GCMS, and found this technique to be far superior to standardized TLC for these particular compounds.
7.7 Application to systematics
The secondary metabolites in over 5000 lichens, approximately 33% of the known species, have now been studied and metabolite data are used more extensively in the routine identification of lichens than in any other group of organisms. These data are not only used extensively in lichen systematics but also in discussions of origins and relationships.
7.7.1 Cortical chemistry For many years taxonomists consistently underestimated the ecolo- gical importance and possible evolutionary significance of the chemical Biochemistry and secondary metabolites 125
Fig. 7.10 Lichen mass spectrum of Lecanora sp. (high mass region). Horizontal scale ¼ m/z values. Vertical scale ¼ % abundance. Molecular ion peaks: A, 4-chloronorlichexanthone; B, 4,5-dichloronorlichexanthone; C, usnic acid; D, arthothelin. components of the upper cortex in lichens. Certainly it was recognized that some cortical substances are correlated with higher taxonomic ranks–for exam- ple, at generic level: vulpinic acid in Letharia; or at the family level: anthra- quinones and particularly parietin in the Teloschistaceae. These cortical compounds appear to have systematic significance because of their vital ecolo- gical roles. In lichens growing on exposed substrates, various light-absorbing com- pounds are located in the upper cortical tissue of the vegetative and generative parts of the thallus, and these cortical lichen substances commonly show varia- tion in concentration along light gradients. Clear evidence suggests that these pigmented compounds have a primary biological role as light-screens, regulat- ing the solar irradiation reaching the algal zone in the upper cortex. In addition to the general filtering effect (Trebouxia grows best at relatively low light intensity), these compounds may have a secondary value in protecting the lichen thallus from excessive ultraviolet irradiation (Solhaug and Gauslaa 1996; Rancan et al. 2002;Rubioet al. 2002). The major groups of substances involved include the -orcinol para-depsides atranorin and chloroatranorin, the usnic acids, anthra- quinones, xanthones and pulvinic acid derivatives. Given their apparent 126 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
physiological importance it seems likely that their formation would have evolu- tionary significance. Indeed, a number of these cortical substances are utilized as correlative characters in the delimitation of genera within the very large family Parmeliaceae (Elix 1993). Interestingly another recent investigation has shown that the species Lecanora somervellii from the high Himalayas produces two effective cortical light-screens, calycin (a pulvinic acid derivative) and usnic acid (a polyketide). This is quite a remarkable adaptation since related species from lower, less-exposed situations produce only cortical usnic acid (Obermayer and Poelt 1992).
7.7.2 Medullary chemistry Variations in medullary constituents are used primarily as discrimina- tors at the species level but also occasionally at generic or suborder level. For example, the distinctive chemical differences between Cetrelia (with orcinol derivatives) and Platismatia (with fatty acids or -orcinol derivatives) are so marked that this character alone is a strong indicator of their generic hetero- geneity (W. Culberson and Culberson 1968). Furthermore, Schmitt and Lumbsch (2004) on the basis of a molecular phylogeny of the Pertusariaceae support secondary chemistry as important systematic characters for the family. At suborder level some chemical substances have restricted distributions. For instance, the hopane triterpenes appear to be distributed in the Lecanor- ineae (e.g. in Physcia, Heterodermia), the Cladonineae (e.g. in some Cladonia), the Teloschistineae (e.g. some Xanthoria) and the Peltigerineae (in Pseudocyphellaria and Peltigera), whereas sesterterpenes are found only in the Peltigerineae (in Pseudocyphellaria) and the dammarane triterpenes in the Lecanorineae (some Pyxine). The use of chemical discriminators at species level has been a controversial topic (Lumbsch 1998). The discovery of chemical differences often led to an appreciation of the importance of previously overlooked morphological features, as in Punctelia subrudecta with lecanoric acid and a pale tan lower surface, and P. borreri with the related tridepside gyrophoric acid and a black lower surface; these species also have different geographic distributions. In some species complexes the morpholological variations may be more subtle, as is observed with the Cladonia chlorophaea group, within which there are correlations of the different chemical races with color and soredia size (at least in parts of their range). Fortunately, most morphologically defined species have a constant chemistry, irrespective of their geographic origin, substrate or ecology, and this justifies the use of chemistry in lichen taxonomy. Within a complex of morphologically similar species, three common patterns of chemical variation are observed: replacement compounds, chemosyndromic variation, and accessory type Biochemistry and secondary metabolites 127 compounds. With replacement type compounds, congeneric chemotypes show simple replacement of one substance by another. Morphologically these lichen populations are sometimes indistinguishable, but they have well-defined, con- stant variations in chemical composition. One of the classical examples is that of Pseudevernia furfuracea, which has three chemical races: an olivetoric acid- containing race from northern Europe; a physodic acid-containing race from southern Europe and north Africa; and a lecanoric acid-containing race from North America. Biogenetically the first two races appear closely related. The metabolites can be considered biosequential, because one can be derived from the other by a single biosynthetic step. But the third race is not related, because lecanoric acid is biosynthetically remote from the other two compounds. It is now generally accepted that, when there is a biogenetic demarcation allied with a biogeographical separation, such taxa should be recognized as species and the North American taxon is distinguished as P. consocians. More recent studies of the European races revealed that they possess distinctive but over- lapping chemistries and show no significant correlation with habitat ecology, a result which convinced some lichenologists that these races represent a single species that shows some chemical variation (Dahl and Krog 1973; W. Culberson et al. 1977). In summary, most lichenologists, who recognize chemically distinct races as species, support their decision primarily on the basis of the different geographic distributions that such races usually show. However, it is suggested by the Culbersons that the best evidence for chemical variation being under genetic control, rather than being environmentally determined, is the fact that chemi- cal races, where sympatric, maintain their integrity even when growing side by side (W. Culberson 1967; W. Culberson and Culberson 1967). As a corollary, the occurrence of chemical intermediates in areas of sympatry either indicates that such races belong to a single species or that hybridization is occurring between the races. The existence of chemosyndromic variation in some lichen groups may make the recognition of chemical intermediates more difficult (C. Culberson and Culberson 1976). A chemosyndrome refers to a group of biosynthetically related metabolites and in this pattern of chemical variation the major meta- bolite (or metabolites) in any one taxon is invariably accompanied by minor quantities of several biosequentially related substances (Table 7.3). Further, the compounds that are the major constituents of some species may be minor constituents of related taxa and vice versa. Hence a true chemical intermedi- ate cannot simply be defined as containing both of two replacement com- pounds, but would have to contain both chemical constellations in comparable concentrations. 128 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Table 7.3. Chemosyndromic variation in the Relicina samoensis complex
Species Fatty (distribution) Echinocarpic Conechinocarpic Hirtifructic Gyrophoric acids Distribution
R. samoensis major minor – – – Pan-Pacific R. amphithrix major minor – – – Australia/ Indonesia R. terricrocodila major minor trace – – Australia R. fijiensis – – major – – Fiji R. niuginiensis – – major trace minor Papua New Guinea R. relicinula – – – – major Indonesia
Source: After Elix (1991).
The biogenetic relationships between secondary products in lichens is used in cladistic analysis of evolutionary relationships amongst taxa. Thus C. Culberson (1986) has used the Cladonia chlorophaea group as an example of how preliminary biogenetic hypotheses can lead to a cladogram for the 14 chemotypes presently known in this complex. A further important feature of chemical races is their ecology. In several cases that are studied in detail, different chemical races were found to be ecologically sorted into distinct habitats in their range of sympatry. Although the underlying physiological causes of this sorting or the related phytogeographically signifi- cant distributions remains unknown, they do indicate that the chemical races have a more than superficial genetic basis. W. Culberson (1986) recently presented a very convincing case for interpret- ing the ecological and biological characteristics of the major chemotypes, into which many Linnean species of lichens are divided, as indicating that these chemotypes are better considered as sibling species rather than as components of traditional morphological species. In fact sibling speciation, where the repro- ductive isolation of populations is often accompanied by ecological but little or no morphological differentiation, is a common product of evolution and is well documented amongst animals and in the vascular plants. Even so, caution must be exercised because of the occurrence of accessory metabolites. These substances occur sporadically in a species, usually in addition to the constant constituents, and have no correlation with any morphological or distributional variations and hence are accorded no taxonomic significance. Such compounds commonly occur as accessory compounds in more than one species and often vary in quantity from deficiency to abundance. Biochemistry and secondary metabolites 129
In summary, most chemotypes (i.e. disregarding accessory chemical varia- tions) appear to have subtle morphological, ecological or distributional tenden- cies and consequently should be afforded some taxonomic recognition.
7.7.3 Cell wall polysaccharides These polymeric storage products of lichens require different techni- ques for their detection, study, and structure determination (Gorin et al. 1988; Common 1991). The best known lichen polysaccharides are lichenan, isoliche- nan, and galactomannan, each of which has a range of different, but related, chemical structures depending on the parent lichen. While secondary products are often useful taxonomically at the specific and generic level in lichens, polysaccharide content is often diagnostic for larger phylogenetic units (Shibata 1973; Common 1991). Polysaccharides have a funda- mental role in the biochemistry of fungi and tend to be conservative features in their evolution. Some polysaccharides are taxonomically significant at the high- est levels of classification. For example, the presence of chitin, chitosan, or cellulose in the cell wall is a feature which helps define the classes of fungi. Within the class Oomycetes, Aronson and coworkers showed that a biochemical dichotomy exists with respect to hyphal wall composition between Rhipidiaceae and Leptomitaceae, the two families comprising the order Leptomitales. Furthermore, these biochemical differences paralleled the traditionally accepted morphological and anatomical differences between these families (Aronson 1977). In a similar vein, Shibata and coworkers (Shibata 1973) showed that pustulan is a characteristic polysaccharide in the Umbilicariaceae (Umbilicaria and Lasallia), and glycopeptides are important cell-wall components of the Lobariaceae. The taxonomic utility of such chemical characters in the Parmeliaceae was developed by Common (1991) who recognized four major groups: (1) isolichenan, (2) a Xanthoparmelia-type lichenan, (3) a Cetraria-type lichenan, or (4) an intermediate-type lichenan. Chemically these polysacchar- ides differ primarily in the stereochemistry of the glycosidic bonds, being largely b in lichenan and a in isolichenan. Xanthoparmelia-type lichenan, Cetraria-type lichenan and the intermediate-type lichenan differ primarily in their staining properties with various iodine reagents (Table 7.4); the structural features responsible for these differences have yet to be elucidated. The utility of this character is readily demonstrated by application to related but well- accepted genera (for example, Hypogymnia contains Cetraria-type lichenan whereas Menegazzia contains isolichenan). It is also used as one of the primary discriminators to differentiate the following yellow-green parmelioid genera: Psiloparmelia and Flavoparmelia (containing isolichenan) from Arctoparmelia (Cetraria-type lichenan) and Xanthoparmelia (Xanthoparmelia-type lichenan). 130 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
Table 7.4. A comparison of the staining properties of isolichenan and various types of lichenan
20 0.15% 0.15% Polysaccharide IKI LPIKI 1.5% IKI CaIKI ZnIKI SIKI Meltzers
Isolichenan blue pale blue bluish bluish bluish – bluish Cetraria-type lichenan – – red deep red – red ppt. orange Xanthoparmelia-type intense – red deep red purple red ppt. deep red lichenan blue Intermediate-type pale blue – red deep red – red ppt. red lichenan
Source: IKI, iodine, potassium iodide solution; LP, lactophenol; S, 10% sulfuric acid; Ca, calcium chloride; Zn, zinc chloride; Meltzers Reagent, chloral hydrate þ iodine potassium iodide. Summarized from Common (1991).
7.8 Application to pharmacology and medicine
The use of lichens in folk medicines persists to the present day (Richardson 1988). Both the Seminole Indians in Florida and the Chinese herbal doctors employ various lichens in medicines, especially as expector- ants. Usnea species were most commonly utilized. Cetraria islandica (‘‘Icelandic Moss’’) is claimed to be effective in treating lung diseases and catarrh, and preparations from this species are still sold in Europe, usually as pastilles. Peltigera canina is eaten in India as a remedy for liver ailments and its high content of the amino acid methionine may be the basis for its alleged curative power (Hale 1983). Today a wide range of secondary metabolites are recognized as having med- icinal value (Pearce 1997). For example, Burkholder et al.(1944; Burkholder and Evans 1945) discovered that extracts from 52 different species of lichen in east- ern North America inhibited growth of several kinds of bacteria. This led to a feverish race to identify the antibiotic components of such lichens. The anti- biotic effect of a number of lichen metabolites was found to be significant for gram-positive bacteria, but ineffective against gram-negative bacteria. Thus gram-positive bacteria were significantly inhibited by usnic acid, protoliches- terinic acid, and a variety of orcinol derivatives. The usnic acids (Ingo´ lfsdo´ ttir 2002) have also been found to exhibit antihis- tamine, spasmolytic, and antiviral properties as well as being active against gram-positive bacteria and streptomycetes. Indeed they are used in commer- cially available antiseptic creams including ‘‘Usno’’ and ‘‘Evosin.’’ Usnic acid is Biochemistry and secondary metabolites 131 reported to be more effective than penicillin salves in the treatment of external wounds and burns and is also used to combat tuberculosis. The active centers of the usnic acid molecule seem to be the benzofuran or dihydrodibenzofuran nucleus, the phenolic hydroxy groups and the 4,4a-double bond in the dihy- droaromatic ring (Asahina and Shibata 1954). The antibiotic action of usnic acid is due to the inhibition of oxidative phosphorylation, an effect similar to that shown by dinitrophenol. More recently Shibuya et al.(1983) showed that 4-O-methylcryptochlorophaeic acid was a powerful inhibitor of prostaglandin biosynthesis and a potentially useful anti-inflammatory drug. Lichen substances are also known to exhibit antitumour activity. Usnic acid has low level activity against lung carcinoma. However, the most active anti- tumor lichen substances are water soluble polysaccharides which appear to be partially O-acetyled homo-D-glucans. The lichen polysaccharide GE-3 was shown to be a host-mediated antitumor active substance, effective because of its stimu- lation of the immune system. The administration of GE-3 to mice caused inflam- matory changes in the liver, a temporary increase in leukocytes followed by excretion of an a1-acid glycoprotein. Purification of the latter led to the isolation of a1-AG-1, which inhibited growth of cancer cells (Shibata 1992). Sulfated GE-3 (GE-3-S) was the most promising agent for HIV suppression. GE-3-S (S content
13.8%, mol. wt. 200 000) was prepared by sulfonation of GE-3 by ClSO3 H. Dextran sulfate and heparin are also suppressive against HIV infection. None of these polysaccharide sulfates expresses an inhibitory effect on cell free HIV reverse transcriptase activity. Therefore the polysaccharide sulfates appear to interfere with the adsorption of HIV particles onto the surface of T4 cells (Shibata 1992). Antifungal activity is also found among lichen substances. Growth of the mould Neurospora crassa is strongly inhibited by usnic acid, as well as by haema- tommic acid, a monocyclic phenol derivative present in some lichen depsides (Hale 1974). A number of lichen metabolites also act as plant-growth regulators, with usnic acid being particularly active (Huneck and Schreiber 1972). Although there are few commercial applications of lichen substances, the variety of antibiotic properties they exhibit obviously encourages further inves- tigations (Pearce 1997; Huneck 1999).
7.9 Harmful properties of lichen substances
In northern Europe the lichen Letharia vulpina was used traditionally as a poison for foxes and wolves. The toxic principle is the pulvinic acid derivative vulpinic acid, which is not only poisonous to all meat eaters but also to insects and mollusks. Surprisingly this compound is ineffective against rabbits and 132 J. A. Elix and E. Stocker-Wo¨ rgo¨ tter
mice. The secalonic acid derivatives are also highly poisonous. These substances are mycotoxins and, like vulpinic acid, may have evolved to serve a twofold ecological role. Thus, in addition to screening incoming light, they are highly poisonous to grazing herbivores. Contact dermatitis, a severe skin rash, is well known among forestry and horticultural workers in North America, forming part of a syndrome known as ‘‘woodcutter’s eczema’’ or ‘‘cedar poisoning.’’ These complaints are an allergic response resulting from exposure to various lichen substances. Among the lichen substances responsible are usnic acid, evernic acid, fumarprotocetraric acid, stictic acid, and atranorin. Usnic acid, for instance, is a common lichen substance in the corticolous species of Alectoria, Evernia, and Usnea, which are widespread in the forests of North America. A dusting of soredia on clothing causes allergic reactions in the wives of lumbermen not directly exposed in the forests. Atranorin and stictic acid are also capable of photosensitizing human skin as well as being contact allergens. This can lead to photocontact dermatitis, where the allergic reactions become much more acute when the persons are exposed to the lichen substances in combination with light (Hale 1983; Richardson 1988). Periodically hundreds of elks die in western North America when these large ruminants are forced out of their normal winter habitats by excessive snows and at lower elevations primarily find Xanthoparmelia chlorochroa to eat (Durrell and Newsom 1939; MacCracken et al. 1983; Anonymous 2004). Although the toxin is fully resolved, the abundance of salazinic acid is suspected. In contrast, these animals eat other epiphytic lichens without apparent ill effects.
7.10 Lichens in perfume
One of the more important economic uses of lichens today is in the perfume industry. The two most important species, Evernia prunastri (‘‘oak moss’’) and Pseudevernia furfuracea (‘‘tree moss’’) are harvested in southern France, Morocco and the former Yugoslavia in large quantities, with a harvest in the range of 8000–10 000 tonnes annually. The combined lichen material and tree bark is subsequently extracted with an organic solvent and treated with ethanol. The concentrate of this solution contains a mixture of essential oils and depside derivatives (degradation products). The final extract with its sweet ‘‘mossy’’ smell is used in some perfumes to ensure persistence on the skin, as the major ingredients do not evaporate readily. The lichen extract may amount to 1–12% of the finished perfume. The precise identity of the scented component remains a trade secret but comprises a very small proportion (c. 0.04%) of the total extract, the majority of which comprises borneol, cineole, geraniol, citronellol, Biochemistry and secondary metabolites 133 camphor, naphthalene, orcinol, orsellinate esters and their homologues (Moxham 1980; Richardson 1988; Hiserodt et al. 2000).
7.11 Lichens in dyeing
Lichens were used as a source of dyestuff from the time of the ancient Greeks and probably earlier (Henderson 1999), but are of little economic impor- tance today. Historically Roccella montagnei, a common fruticose lichen on rocks, provided valuable red or purple dyes in the Mediterranean region. These dyes were produced by ‘‘fermenting’’ the Roccella or chemically equivalent species (Ochrolechia tartarea, O. androgyna,orParmotrema tinctorum) with dilute ammonia solution. The macerated lichen and dilute ammonia were sealed in a container containing twice the volume of air. The purple color developed after a week and was used as a direct dye (orchil) for protein fibres (wool and silk). The simple para-depsides erythrin (Roccella) and lecanoric acid (Ochrolechia and Parmotrema tinctorum) present in these lichens are responsible for these colors. Rapid base hydrolysis of the lecanoric acid or erythrin by ammonia gives ammonium orsellinate and then orcinol (by decarboxylation). Subsequent oxidative cou- pling in the presence of ammonia gives rise to the dyestuff, orcein, which comprises a mixture of three major chromophores, 7-hydroxyphenoxazone, 7-aminophenoxazone and 7-aminophenoxazine (Hale 1983). The common acid-base indicator litmus, formerly widely used in chemistry laboratories, is closely related to orcein but represents a more complex mixture of polymeric compounds with the 7-hydroxyphenoxazone chromophore and its anion being responsible for the sensitivity of the color to pH. Some Harris tweeds manufactured in Scotland are still dyed with lichen dyestuffs (Richardson 1988). All of these dyes are quite colorfast, and impart a unique musty odor to the fabric. For instance, Parmelia omphalodes is utilized to provide a rich brown dye for dyeing protein fibres, particularly wool. Subsequent investigations showed that this was due to the salazinic acid pre- sent, and in fact most of the tawny yellow-brown to reddish-brown colors produced on wool by lichen dyes are produced by lichen substances with o-hydroxyaldehyde functionalities. The aldehyde functional group condenses with free amino groups present in the wool proteins to form a stable Schiff base (azomethine) linkage (Hale 1974). For practical applications, Casselman (2001)is a useful reference. 8 Stress physiology and the symbiosis
r. p. beckett, i. kranner, and f. v. minibayeva
Lichens are the dominant life forms in about 8% of the land surface of the Earth (Ahmadjian 1995), mainly in polar regions and on the tops of moun- tains. These places are characterized by severe abiotic stresses such as desicca- tion, temperature extremes, and high light intensities. Arguably, what really makes lichens special, and what separates them from most other eukaryotic organisms, is their ability to tolerate extreme stresses. For this reason, some have called lichens ‘‘extremophiles,’’ organisms that can thrive in conditions that would kill other, less specialized organisms. Scientists have found that hardy lichens can survive a trip into space, and now the list of natural astronauts includes lichens. During a recent experiment by the European Space Agency, lichen astronauts were placed on board a rocket and launched into space, where they were exposed to vacuum, extreme temperatures, and ultraviolet radiation for two weeks. Upon analysis, it appeared that the lichens handled their space- flight just fine (Young 2005)! In the typical environments that many lichens inhabit, stresses such as low thallus water content and temperature extremes can develop within just a few minutes. However, others, such as a nutrient deficiency, can take months to develop. The stressfulness of a particular habitat is the result of the interaction of climate and substrate. It plays a major role in determining lichen distribution. Understanding the physiological processes that lie behind stress injury, and how lichens tolerate environmental stress, is therefore of great importance in lichen biology. A ‘‘stress’’ factor can be defined as any external influence that has a harmful effect on an organism. This chapter will discuss environmental or ‘‘abiotic’’ factors that produce stress in lichens, although biotic factors, such as competing higher plants or other lichens, pathogens, and insect predation can also result in
Lichen Biology, ed. Thomas H. Nash III. Published by Cambridge University Press. # Cambridge University Press. Stress physiology and the symbiosis 135 stress. It is rarely possible to see at a glance whether a lichen is alive or dead. Furthermore, because of their slow growth rates, it is difficult to use growth to assess stress. Instead, lichenologists tend to measure parameters such as the inhibition by stress of net photosynthesis or chlorophyll fluorescence to quan- tify stress effects on the photobiont. If researchers are more interested in the mycobiont, they may study the stress-induced inhibition of respiration or leak- age of intracellular soluble potassium through membranes. The concept of stress is closely linked to that of stress tolerance, which is the ability of an organism to cope with an unfavorable environment. If tolerance increases as a result of exposure to prior stress, the plant is said to be acclimated (or hardened). Acclimation can be distinguished from adaptation, which usually refers to a genetically determined level of resistance acquired by a process of selection over many generations. Adaptation and acclimation to environmental stresses result from changes that occur at all levels of organization, from the anatomical and morphological level to the cellular, biochemical, and molecular level. The specific stresses we will consider here include desiccation, tempera- ture extremes, and high light intensities. Air pollution is an important source of stress, and is discussed in Chapter 15, while Chapters 11 and 12 review nutrient stresses. Although it is convenient to examine each of these stresses separately, most are interrelated, and a common set of cellular, biochemical, and molecular responses accompanies many of the individual acclimation and adaptation processes. For example, desiccation and light stress often accompany high temperatures. Furthermore, at the cellular level, many stresses may have the same effects, for example the production of reactive oxygen species and damage to the cytoskeleton. We shall see that while each stress causes particular pro- blems for lichens, the underlying mechanisms of resistance probably share many common features with each other, and with those in other organisms.
8.1 Stress tolerance – protection or repair?
We are still far from understanding fully the mechanisms of stress tolerance in lichens. In particular, we are not sure if tolerance is largely a matter of reassembly and reactivation of components conserved intact through a time of stress, or whether a more or less extensive ‘‘repair’’ process is involved. For example, during rehydration following desiccation some parameters, for exam- ple chlorophyll fluorescence, recover almost immediately following rehydra- tion (e.g. Beckett et al. 2005b). This suggests that the integrity of the thylakoids is preserved throughout the events of desiccation and remoistening. On the other hand, the fact that complete recovery of carbon fixation can take much longer suggests the involvement in recovery of other cellular mechanisms, as yet 136 R. P. Beckett et al.
unknown. Furthermore, during rehydration following desiccation, membranes are initially leaky to ions and metabolites, but later regain their integrity (Weismann et al. 2005a). Therefore, it is possible that some form of repair- based desiccation tolerance mechanism exists. In bryophytes, initial studies on moss ultrastructure during and after desiccation seemed to imply that some form of repair takes place. Freshly rehydrated plants appear to display swelling of chloroplasts and mitochondria, and major changes in the endomem- brane domains and microtubular cytoskeleton – damage that is repaired gradu- ally. However, recent more careful investigations using improved procedures for preparing material for microscopy (e.g. Pressel et al. 2006; Proctor et al. 2006) no longer support a simple damage-repair hypothesis of desiccation tolerance. The latest view is that tolerance involves a suite of protective mechanisms, including scavenging reactive oxygen species, or preventing their formation, and probably synthesizing sugars and dehydrins. The latter are LEA (late embryo- genesis abundant)-like proteins that share some features with LEA proteins in seeds. In bryophytes, recovery of the essential systems, such as respiration, light capture and carbon dioxide fixation, and protein synthesis, now looks to be largely physical, and probably not metabolically costly in terms of either energy or materials. We need to do more work to test if the same is true for lichens. If so, then the ‘‘cost’’ of a lichen being so stress tolerant may well be mainly in producing ‘‘protective mechanisms’’ that enable lichens to survive the next stress event.
8.1.1 Limits to stress tolerance Early workers determined the limits of stress tolerance in lichens, and Kappen (1974) provides an excellent review of this work. The main conclusions of this review are summarized here. For water supply, the great majority of lichens are highly desiccation tolerant. Providing that desiccation occurs rea- sonably slowly (over hours rather than minutes), most lichens can withstand drying to water contents of 5% or less, and most can remain viable for months, if stored at low relative humidities. Even aquatic species such as Dermatocarpon fluviatile can survive desiccation for four weeks. Conversely, most lichens are highly intolerant of submergence, or in many cases even moist storage, for more than a few days. Such lichens appear to become overrun with pathogenic fungi, or dissociate into separate symbionts. For temperature, lichens are tolerant of extremely low temperatures (e.g. liquid nitrogen, 196 8C) when dry. Hydrated thalli can also tolerate these temperatures, as can even tropical species, provid- ing that cooling is slow enough. Interestingly, the heat tolerance of hydrated lichens is lower than that of higher plants, and most lichens, including tropical species, die when the thallus temperature exceeds 35 to 43 8C. Normally, in the Stress physiology and the symbiosis 137 field, lichens will dry before they reach these temperatures. However, even in temperate climates, the temperature of dry thalli can reach 60 8C – not surpris- ingly, as the heat tolerance of dry thalli is high. For example, Teloschistes flavicans easily survived three days at 60 8C. Generally, lichens growing in open habitats have a higher tolerance to high temperatures. More recently, high light stress has received much attention. Although some species possess carbon concentrating mechanisms (see Chapters 9 and 10), lichens display classical C3 carbon fixation and therefore light intensities will often exceed those that saturate photosynthesis. A significant decrease in the levels of stratospheric ozone has been observed during the last few decades, with accompanying increases of UV radiation. It is unclear if hydrated lichens are more tolerant to UV radiation than higher plants. However, some lichens can grow in very low light intensities in caves, where apparently they must be deriving carbon saprophytically. Finally, because lichens grow in exposed sites, they must be subjected to considerable mechanical damage from forces such as wind and abrasion by sand particles. The ability of lichens to propagate by ‘‘thallus fragmentation’’ (see Chapter 4) suggests that they have strong tolerance to mechanical stress.
8.2 Harmful effects of stress
As already mentioned, each type of stress can potentially cause specific problems for lichens. For example, desiccation damages the cytoskeleton, makes membranes leaky, and changes the structure of proteins so that, for example, the activity of some enzymes is reduced. Chilling is responsible for a different kind of membrane damage, a large increase in viscosity called ‘‘gel- ling’’, while heat can denature proteins. It seems unlikely therefore that we will ever have a concept of a ‘‘general resistance of the protoplasm’’ that we can apply to lichens. However, one feature shared by almost all stresses is that they cause the formation of reactive oxygen species (ROS). Intracellularly produced ROS can cause considerable damage to cells by attacking nucleic acids, lipids, and proteins. To survive stress, lichens must be able to either reduce the forma- tion of ROS, or detoxify them once formed.
8.2.1 Formation of ROS Many ROS are free radicals, atoms, or molecules with unpaired elec- trons. This unpaired electron is readily donated, and, as a result, most free radicals are highly reactive. Oxygen radicals include superoxide (O2 ), the hydroxyl radical ( OH), hydroperoxyl (protonated superoxide, HO2 ) and the 1 nitric oxide radical (NO ). Hydrogen peroxide (H2O2), singlet oxygen ( O2), 138 R. P. Beckett et al.